multiple sites of extinction for a single learned...

doi:10.1152/jn.00381.2011 107:226-238, 2012. First published 21 September 2011;J NeurophysiolBrian E. Kalmbach and Michael D. MaukresponseMultiple sites of extinction for a single learned

You might find this additional info useful...

62 articles, 28 of which can be accessed free at:This article cites http://jn.physiology.org/content/107/1/226.full.html#ref-list-1

including high resolution figures, can be found at:Updated information and services http://jn.physiology.org/content/107/1/226.full.html

can be found at:Journal of Neurophysiologyabout Additional material and information http://www.the-aps.org/publications/jn

This infomation is current as of January 4, 2012.

American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/.(monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2012 by the

publishes original articles on the function of the nervous system. It is published 12 times a yearJournal of Neurophysiology

on January 4, 2012jn.physiology.org

Dow

nloaded from

http://jn.physiology.org/content/107/1/226.full.html#ref-list-1

http://jn.physiology.org/content/107/1/226.full.html

http://jn.physiology.org/

Multiple sites of extinction for a single learned response

Brian E. Kalmbach1 and Michael D. Mauk1,2

1Center for Learning and Memory, 2Section of Neurobiology, University of Texas, Austin, Texas

Submitted 25 April 2011; accepted in final form 19 September 2011

Kalmbach BE, Mauk MD. Multiple sites of extinction for asingle learned response. J Neurophysiol 107: 226 –238, 2012. Firstpublished September 21, 2011; doi:10.1152/jn.00381.2011.—Mostlearned responses can be diminished by extinction, a process that canbe engaged when a conditioned stimulus (CS) is presented but notreinforced. We present evidence that plasticity in at least two brainregions can mediate extinction of responses produced by trace eyelidconditioning, where the CS and the reinforcing stimulus are separatedby a stimulus-free interval. We observed individual differences in theeffects of blocking extinction mechanisms in the cerebellum, thestructure that, along with several forebrain structures, mediates acqui-sition of trace eyelid responses; in some rabbits extinction was pre-vented, whereas in others it was largely unaffected. We also show thatcerebellar mechanisms can mediate extinction when noncerebellarmechanisms are bypassed. Together, these observations indicate thattrace eyelid responses can be extinguished via processes operating atmore than one site, one in the cerebellum and one upstream inforebrain. The relative contributions of these sites may vary fromanimal to animal and situation to situation.

Pavlovian; nictitating; eyeblink; plasticity; learning; delay; trace;cerebellum; prefrontal cortex

BEHAVIORAL EXTINCTION is an important aspect of learning.Learned responses that were once appropriate may becomeinappropriate in other contexts or as circumstances change, andthe mechanisms of extinction are relevant to a number ofpsychiatric disorders (Peters et al. 2009; Phelps and LeDoux2005). As with the mechanisms of acquisition, identifying thesites of plasticity is an essential initial step toward understand-ing the mechanisms of extinction. Eyelid conditioning is anexample of the many forms of learning that require more thanone site of plasticity (Medina et al. 2002b). For example, delayeyelid conditioning involves plasticity in at least two siteswithin the cerebellum (Ohyama et al. 2006; Perrett et al. 1993).Such instances raise interesting and potentially complex pos-sibilities for the site(s) of plasticity underlying extinction. Ifplasticity at more than one site is necessary for the expressionof learned responses, reversing or counteracting plasticity atany site may be sufficient to mediate extinction. Moreover,multiple sites of plasticity, each potentially sufficient to sup-port extinction, create the potential for animal-to-animal andsituation-to-situation variability in the site or sites that mediateextinction. We report evidence that at least two different sitesof plasticity can contribute to the extinction of trace eyelidresponses in rabbit.

Eyelid conditioning provides many advantages for address-ing the mechanisms of extinction. Delay eyelid conditioning,where pairing a conditioned stimulus (CS; often a tone) with areinforcing unconditioned stimulus (US) promotes learned eye-

lid closure to the CS, is mediated by plasticity in both thecerebellar cortex and deep cerebellar nuclei (DCN) (Medina etal. 2001; Ohyama and Mauk 2001; Ohyama et al. 2006; Perrettet al. 1993). This plasticity is engaged by pairing tone-drivenmossy fiber inputs (Halverson and Freeman 2010; Steinmetz etal. 1987) with US-driven climbing fiber inputs to the cerebel-lum (Mauk et al. 1986; Rosen et al. 1989; Yeo et al. 1986).Evidence indicates that delay responses extinguish via plastic-ity in cerebellar cortex, leaving plasticity in the DCN some-what intact (Fig. 1) (Jirenhed et al. 2007; Medina et al. 2001,2002a; Perrett and Mauk 1995).

Although delay eyelid conditioning is largely independent offorebrain structures (Mauk and Thompson 1987; Oakley andRussell 1972), trace eyelid conditioning, in which a stimulus-free trace interval separates CS offset and US onset, requiresseveral forebrain structures for trace intervals longer than�300–400 ms and engages cerebellar learning via a mossyfiber input driven by medial prefrontal cortex (mPFC) inaddition to the tone-driven mossy fiber input necessary fordelay conditioning (Fig. 1) (Fontan-Lozano et al. 2005;Kalmbach et al. 2009, 2010a, 2010b; Mauk and Thompson1987; Moyer et al. 1990; Powell and Churchwell 2002; Ryouet al. 2001; Takehara et al. 2003; Weible et al. 2000, 2007).This highlights the possibility that both cerebellar and fore-brain sites of plasticity could support extinction in trace eyelidconditioning.

We report evidence that the extinction of trace eyelid re-sponses can indeed be mediated by more than one site, one inthe cerebellum and another presumably in upstream forebrainstructures. We also found evidence that the site of extinctioncan vary from animal to animal. These results demonstrate thatextinction of a single learned response can be mediated byplasticity at more than one site. The contributions from thesesites apparently can vary depending on the situation or learningregimen, and this may even explain the bimodal distribution ofextinction rates observed during extinction.

MATERIALS AND METHODS

Subjects and surgery. Male New Zealand albino rabbits (Orycto-lagus cuniculus; Myrtle’s Rabbitry, Thompsons Station, TN) initiallyweighing 2.5–3 kg served as subjects. Rabbits were individuallyhoused, maintained on a fixed daily diet, and given free access towater. All surgical and experimental procedures were approved byThe University of Texas at Austin Institutional Animal Care and UseCommittee and were in accordance with the National Institutes ofHealth guidelines.

Rabbits were prepared for microstimulation and/or infusion exper-iments using sterile surgical procedures. An initial anesthetic cocktailof ketamine (40 mg/kg) and acepromazine (5 mg/kg) was deliveredvia subdermal injection. Rabbits were then placed in a stereotaxicrestrainer, and general anesthesia was maintained with isofluorene(2% mixed in oxygen) for the remainder of the surgery. An incisionof �4 cm was made along midline, and the skin was retracted to

Address for reprint requests and other correspondence: B. E. Kalmbach,Center for Learning and Memory, Section of Neurobiology, Univ. of Texas, 1Univ. Station, C7000, Austin, TX 78712 (e-mail: [email protected]).

J Neurophysiol 107: 226–238, 2012.First published September 21, 2011; doi:10.1152/jn.00381.2011.

226 0022-3077/12 Copyright © 2012 the American Physiological Society www.jn.org


Dow

nloaded from


reveal the lambda and bregma landmarks of the skull. A 3-mm-diameter craniotomy was then drilled to accommodate the cannulaand/or stimulation electrodes, and four other smaller craniotomieswere drilled to accommodate anchor screws. Lambda was then posi-tioned 1.5 mm ventral to bregma. In six subjects, a stainless steelguide cannula (Plastics One, Roanoke, VA) was targeted for the rightdorsal accessory olive (1.0 mm anterior, 0.7 mm lateral, and 22 mmventral from lambda). In eight subjects, a guide cannula was targetedfor the left anterior interpositus nucleus of the DCN (1 mm anterior,5 mm lateral, and 13.3 mm ventral from lambda). In five subjects, twoclosely spaced tungsten stimulating electrodes laterally separated by1.0 mm (A-M Systems, Carlsborg, WA; �100–200 k�) were placedin the left middle cerebellar peduncle (3.0 mm anterior, 5.0 mmlateral, and 16.0 mm ventral), and a guide cannula was placed in theinterpositus nucleus. In these subjects, a stainless steel screw placedon the surface of the brain served as ground. In all subjects, including21 subjects used only for behavioral experiments, a bolt was placedbetween the skull screws and dental acrylic was used to secure theimplants in place. Loose skin was sutured, and a dummy cannula wasplaced in the guide cannula. Finally, stainless steel electrodes wereplaced at the margins of the left eye (one rostral and the other caudal)to deliver the electrical stimulation US. Rabbits were given postop-erative analgesics and antibiotics for 2 days and were allowed torecover for a week before experiments began.

Conditioning. Subjects were trained in custom-designed, well-ventilated, and sound-attenuating chambers measuring �90 � 60 �60 cm (length, width, height). To generate tones for the CS andelectrical pulses for the US, each chamber was equipped with aspeaker connected to an audio source module (model V85-05; Coul-born Instruments, Allentown, PA) and with isolated pulse stimulators(model 2100; A-M Systems, Carlsborg, WA) connected via leads tothe electrodes implanted around the eye. Other stimulus timers wereused to time the delivery of constant-current pulses by stimulusisolators (model 2300; A-M Systems) through leads connected toelectrodes implanted in the middle cerebellar peduncle. The positionof the left external eyelid was measured by using an infrared emitter/detector to detect changes in the amount of reflected light as theeyelids open and close. At the start of each daily conditioning/testsession, the eyelid detector was attached to the bolt on the head stageof each rabbit and calibrated by delivering the US to elicit maximumeyelid closure. The amplitude of the signal was adjusted to match anassumed 6-mm maximum eyelid closure.

Stimulus presentation and data acquisition were controlled bycustom-designed software that was run on a computer adjacent to theconditioning chambers. Subjects received trace conditioning, dualdelay/trace conditioning, or conditioning with mossy fiber stimulationas the CS. All daily conditioning sessions consisted of 12 blocks of 9trials where the first trial of each block was the CS presented alone.Trials were separated by a mean interval of 30 s (�10 s).

Twenty-one rabbits were trained for 10–12 days in trace condi-tioning with a 1-kHz sinusoidal tone (ramped at onset and offset witha time constant of 5 ms to avoid clicks). The US was a 50-ms train ofconstant-current pulses (50 Hz, 1-ms pulse width, 2–3 mA) deliveredthrough electrodes implanted around the left eye. For paired traceconditioning trials, the CS was presented for 500 ms and was followed500 ms later by the US.

Fourteen rabbits were trained for 10–15 days using a dual delay/trace conditioning paradigm in which delay and trace conditioningtrials alternated. The tone (1 or 9.5 kHz) used to signal each trial typewas counterbalanced across subjects. Trace conditioning trials werethe same as described above, whereas for paired delay conditioningtrials the 50-ms US was presented 500 ms into a 550-ms CS.

Five rabbits were trained with electrical stimulation of mossy fibersas the CS. Two stimulus trains with durations that approximatetone-driven (Aitkin and Boyd 1978; Boyd and Aitkin 1976; Campo-lattaro et al. 2011; Freeman and Muckler 2003) and putative mPFC-driven mossy fiber inputs during trace conditioning (Takehara-Nishiuchiand McNaughton 2008) with a 500-ms CS and 500-ms trace intervalwere delivered through two separate electrodes implanted in themiddle cerebellar peduncle (where mossy fibers enter the cerebellum).This pattern of input has previously been shown to be necessary andsufficient to produce responses that quantitatively match trace condi-tioned responses to a tone (Kalmbach et al. 2010a). Stimulus trainsconsisted of 100-�s, 100-�A constant-current pulses delivered at 50Hz. The tone-mimicking input lasted for 500 ms, and the mPFC-mimicking input lasted for 1,000 ms. Training sessions were other-wise identical to those described above. Rather than train subjects fora set number of days, subjects were trained one session past thesession in which they first reached a 60% response rate. We stimulatedmossy fibers rather than their cell bodies in the pontine nuclei tominimize activation of fibers of passage or antidromic activation ofpontine-projecting forebrain regions. All electrodes were placedwithin the middle cerebellar peduncle �2–3 mm anterior to theanterior interpositus nucleus as confirmed by histological analysis.Furthermore, stimulation through each electrode did not elicit anynoticeable movements. Extinction sessions were identical to trainingsessions except that all trials were CS-alone presentations.

Test sessions and infusions. After these initial training sessions, asubset of subjects were given extinction sessions where infusionswere made in either the DCN (8 dual delay/trace-trained and 5stimulation-trained subjects) or the dorsal accessory olive (6 dualdelay/trace-trained subjects). Drugs were dissolved in artificial cere-brospinal fluid (ACSF) consisting of (in mM) 124 NaCl, 3.0 KCl, 26.0NaHCO3, 1.3 NaH2PO4·H2O, 2.0 MgCl, 10.0 dextrose, 10.0 HEPES(pH 7.35), and 2.0 CaCl2. Infusions were made through a 33-gaugeinfusion cannula that extended 1.2 mm beyond the surgically im-planted guide cannula. The infusion cannula was coupled to a 50-�lHamilton syringe that was mounted on an automated injector system

Fig. 1. The pathways known to mediate the acquisition and extinction of delay (black) and trace (black and gray) eyelid conditioning. Acquisition of delayconditioning involves activation of mossy fiber inputs to the cerebellum by the tone and activation of climbing fibers originating in the inferior olive by theunconditioned stimulus (US). Trace conditioning involves an additional mossy fiber input driven by prefrontal cortex. Extinction of delay eyelid responsesrequires inhibition of climbing fibers via output from the cerebellum. Extinction of delay responses can be blocked by silencing deep nucleus neurons (a) andblocking the GABAergic synapses in the inferior olive (b). Extinction can be caused during normal training by blocking the excitatory synapses in the inferiorolive (c).

227MULTIPLE EXTINCTION SITES

J Neurophysiol • doi:10.1152/jn.00381.2011 • www.jn.org


Dow

nloaded from


(model MD-1001; Bioanalytical Systems, West Lafayette, IN) anddriven by an electronic pump (model MD-1020).

In subjects with the cannula in the DCN (8 tone-trained and 5stimulation-trained subjects), there were two infusion sessions(Fig. 2). Twenty minutes before the start of the extinction session,ACSF or the GABA agonist muscimol (1 mM; Tocris) was infusedinto the DCN at a rate of 0.1 �l/min (total volumes 7–8 �l). Theinfusion then continued for the remainder of the session at the samerate. An additional day of extinction was given after each infusionsession, and subjects were retrained between infusion sessions (eachsubject received both types of infusions on separate days). Forstimulation-trained subjects, the infusion session administered firstwas counterbalanced across subjects.

Subjects with the cannula in the inferior olive were administeredtwo types of infusion sessions. Twenty minutes before the start of theextinction session, ACSF or the GABAA antagonist gabazine (20 �M;Tocris) was infused into the inferior olive at a rate of 0.1 �l/min (totalvolumes 7–8 �l). The infusion then continued for the remainder of thesession at the same rate. All subjects were given each type of infusion,and the type of infusion given first was counterbalanced acrosssubjects. If gabazine infusions did not affect delay responding, sub-jects were retrained and the infusion session was repeated with a0.5-mm longer infusion cannula. Two subjects were given an addi-tional infusion session where the AMPA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(f)quinoxaline (NBQX; 150 �M) wasinfused into the olive at a rate of 0.1 �l/min. In one subject, theinfusion began 20 min before the start of the session, and in the othersubject, the infusion began 40 min before the start of the session. Eachinfusion continued for the remainder of the session at a rate of 0.1�l/min.

Data analysis. Eyelid responses during each trial were digitized (1kHz, 12-bit resolution) and stored on disk for subsequent off-lineanalyses using custom software. For each response, 2,500 points wererecorded, the 200 ms before CS onset and the 2,300 ms that followed.Trials in which eyelid movements �0.3 mm were made in the 200 msbefore CS onset were automatically excluded from analysis by thesoftware (fewer than 5%). A conditioned response during pairedconditioning sessions was defined as an eyelid movement of at least0.3 mm within the interval between CS onset and US onset (inter-stimulus interval; ISI). For extinction training, this interval wasextended 200 ms beyond the training ISI. Latency to onset wasdetermined using an algorithm designed to detect the initial deflectionof each response away from the baseline. The criterion for extinctionwas defined as eight non-conditioned responses in nine consecutivetrials. Thus trials to criterion is the total number of trials performed toreach this criterion.

Repeated-measures analysis of variance (ANOVA) followed byF-tests for simple effects and Tukey’s post hoc comparisons wereused to test for within-subject differences. The Kolmogorov-Smirnov

test was used to test for normality. All tests were two tailed with asignificance level of 0.05.

Histology. After the completion of experiments, marking lesionswere made by passing DC anodal current (200 �A) for 20 s throughthe electrodes or a stainless steel wire cut the length of the guidecannula. Animals were killed with an overdose of pentobarbitalsodium and perfused transcardially with 1 liter of 0.9% saline fol-lowed by 1 liter of 10% formalin. Brains were extracted and stored in10% formalin for at least a week. They were then embedded in analbumin gelatin mixture and sectioned using a freezing microtome(80-�m sections). Tissue was mounted and stained with cresyl violet.

RESULTS

Disrupting cerebellar mechanisms of extinction preventsextinction of delay but not trace responses. Previous work hasshown that inhibition of the climbing fiber input to the cere-bellum via a projection from DCN to the inferior olivary nuclei(Bengtsson et al. 2004; Hesslow and Ivarsson 1996) is thenecessary and sufficient signal for extinction of delay eyelidresponses (Medina et al. 2002a). Different methods of blockingthis inhibitory signal all prevent the extinction of delay eyelidresponses (Hardiman et al. 1996; Medina et al. 2002a; Ram-nani and Yeo 1996). We began by testing whether blockingthis inhibitory signal, and thus blocking the cerebellar mecha-nisms of extinction, prevents extinction of trace eyelid re-sponses. We employed both methods known to prevent extinc-tion of delay eyelid responses: infusing a GABAA receptoragonist into the DCN to block activity of DCN neurons (Fig.1a) (Hardiman et al. 1996; Ramnani and Yeo 1996) andinfusing a GABAA receptor antagonist into the inferior olive(Fig. 1b) (Medina et al., 2002a). We implemented these exper-iments in rabbits trained with both delay and trace conditioningto permit within-animal comparison of the effects on extinctionof delay and trace responses.

Preventing cerebellum-mediated extinction by silencingneurons in DCN. Silencing DCN neurons with infusions of theGABAA receptor agonist muscimol not only prevents DCN-mediated inhibition of the climbing fibers during the CS butalso prevents the expression of conditioned responses (Chap-man et al. 1990; Garcia and Mauk 1998; Hardiman et al. 1996;Kalmbach et al. 2009; Ramnani and Yeo 1996). Because ofthis, the ability of a muscimol infusion to block extinction mustbe inferred from levels of responding during the postmuscimolextinction session the next day (Ramnani and Yeo 1996). Inessence, if responding during this second postmuscimol extinc-

Fig. 2. A: timeline of the experiment (top) used to deter-mine whether inactivation of deep cerebellar nuclei (DCN)prevents the extinction of delay and trace eyelid responses.Although extinction is potentially blocked (bottom) duringthe infusion of muscimol on the first day of extinction(extinction 1), this effect can only be detected by compar-ing responding on the next extinction session postmuscimol(extinction 2) with the 2 days of control extinction (control1 and control 2) as shown in B. No effect on extinction isindicated by similar performance during postmuscimol andcontrol 2 sessions (and thus a significant difference be-tween postmuscimol and control 1 sessions). In contrast,similar performance between postmuscimol and control 1sessions (and thus a significant difference between post-muscimol and control 2 sessions) indicates that extinctionwas prevented by the infusion. ACSF, artificial cerebralspinal fluid; CR, conditioned response.

228 MULTIPLE EXTINCTION SITES



Dow

nloaded from


tion session is similar to that during a typical first session ofextinction, then the previous day’s infusion of muscimolblocked extinction. In contrast, lower levels of respondingmore typical of a second session of extinction training indicatethe infusion did not block extinction (Fig. 2A).

To make these comparisons for both delay and trace condi-tioning, we trained eight rabbits using a dual delay/traceconditioning procedure until an asymptotic level of respondingwas reached. In this procedure, one tone CS was used withdelay conditioning and a second tone CS was used with traceconditioning, and for all analyses the delay and trace trialswere analyzed separately for each animal. Each rabbit was thengiven a standardized set of five sessions (Fig. 2A): 1) extinctionduring infusion of muscimol to block cerebellar mechanisms ofextinction, 2) a postmuscimol extinction session, 3) retrainingback to asymptotic performance, 4) re-extinction during acontrol infusion of the ACSF vehicle (control day 1), and 5) asecond (control day 2) extinction session with no infusion. Thelatter 2 days (control 1 and control 2) serve as the controllevels of responding on the first and second days of normalextinction. Again, the key comparisons are the rates of extinc-tion on the postmuscimol session vs. the control 1 session andvs. the control 2 session. These within-animal comparisons arepossible, because previous studies have shown that repeatedrounds of extinction and reacquisition do not appreciablychange the rate of trace response extinction (Kehoe 2006).

It is essential to the interpretation of this experiment that theinfusions successfully abolished the expression of conditionedresponses during the muscimol infusion session. To test theeffectiveness of these infusions, we compared the rate ofresponding during the muscimol infusion session to the rate ofresponding during the later ACSF infusion session. Figure 3shows that the infusion of muscimol into the DCN abolished

both delay [32.7 � 5.0% responding during ACSF infusion,5.6 � 2.6% responding during muscimol infusion; t(7) � 5.26,P � 0.001] and trace eyelid responses [12.86 � 2.2% respond-ing during ACSF infusion, 3.5 � 1.1% responding duringmuscimol infusion; t(7) � 4.46, P � 0.003]. Sample cannulaplacements in the DCN are shown in Fig. 3B.

Comparison of performance during the subsequent postmus-cimol extinction session with the control 1 and control 2sessions revealed that the muscimol infusions prevented theextinction of delay but not trace eyelid responses. For delayconditioning, the level of responding during the entire post-muscimol extinction session was indistinguishable from thatduring the control 1 session (P � 0.63; paired t-test) and wassignificantly greater than that during the control 2 session[t(7) � 4.58, P � 0.003; paired t-test]. We also examined therate of extinction for these groups with a two-way repeated-measures ANOVA. Figure 4A shows that delay responsesduring postmuscimol and control 1 sessions extinguished atsimilar rates (compare dark gray and light gray curves in Fig.4A; P � 0.32; block � session interaction effect), whereasresponses during the control 2 session extinguished morequickly [black curve; F(11, 77) � 2.27, P � 0.02 for postmus-cimol and F(11, 77) � 6.15, P � 0.001 for ACSF (control 1);block � session interaction effects]. Thus, in terms of overallresponding for the entire session and in terms of rate ofextinction, the postmuscimol delay responding was like a first(control 1) day of extinction and not like a second (control 2)day of extinction. These results indicate that, for the delayresponses, the process of extinction was blocked during themuscimol infusion. Conversely, for the trace responses, themuscimol infusion, control 1, and control 2 sessions did notdiffer in terms of the overall level of responding during theentire session, but the postmuscimol and control 1 sessionsshowed significantly different rates of extinction (Fig. 4, A andC). Responses during the postmuscimol and control 2 sessionsextinguished at the same rate (compare dark gray and blackcurves in Fig. 4A; P � 0.94 for block � session interactioneffect), whereas extinction during the postmuscimol sessionoccurred more quickly than for the control 1 session [light graycurve; F(11, 77) � 5.02, P � 0.001 for postmuscimol block �session interaction effect]. Thus the postmuscimol trace re-sponding was different from a first (control 1) day of extinctionand indistinguishable from a normal second (control 2) day ofextinction. These results indicate that the muscimol infusionhad no measurable effect on the extinction of the traceresponses.

This differential effect of muscimol infusions on delay vs.trace extinction is clearly apparent in the session differencescores obtained by subtracting (for each animal) the responselikelihood of each training block during the postmuscimolextinction session from the same block of either the control 1session or the control 2 session. A score of zero indicates thatperformance during the two sessions under comparison is equalfor a given training block. For delay conditioning, responselikelihood during postmuscimol sessions was approximatelyequal to control 1 sessions (difference scores � 0), whereasresponse likelihood during postmuscimol extinction wasgreater than control 2 extinction for the first few trainingblocks (difference scores � 0; Fig. 4B, left). For trace condi-tioning, the inverse pattern of results is apparent; responselikelihood during postmuscimol and control 2 extinction was

Fig. 3. A: DCN inactivation prevented the expression of delay and traceresponses. **P � 0.01, %CR during control 1 vs. %CR during muscimolinfusion (paired samples t-test). B: sagittal sections of cerebellum show 2sample cannula placements (arrows). C: point-by-point averages of eyelidposition sweeps of 8 subjects during the muscimol infusion session. Blackshaded regions denote the presence of the conditioned stimulus (CS), whereasdark gray shaded regions denote the trace interval used in paired trails. Trialsare arranged from first on the bottom to last on top. An upward deflectionindicates the animal closed its eye. Inactivation of DCN abolished the expres-sion of delay and trace eyelid responses.




Dow

nloaded from


approximately equal across training blocks (difference scores� 0), whereas response likelihood during postmuscimol ex-tinction was less than the control 1 session for the first fewtraining blocks (difference scores � 0; Fig. 4B, right). To-gether, these data suggest that inactivation of DCN, and thusblocking cerebellum-mediated inhibition of climbing fibers,prevents the extinction of delay eyelid responses, which repli-cates the findings of previous studies (Hardiman et al. 1996;Ramnani and Yeo 1996), but does not prevent the extinction oftrace eyelid responses.

Preventing cerebellum-mediated extinction by blocking in-hibition of climbing fibers. Blocking the inhibition of climbingfibers directly via infusion of the GABAA receptor antagonistgabazine into the inferior olive (Medina et al. 2002a) repre-sents a second and more direct way to interrupt extinctionmechanisms in the cerebellum. This procedure has the advan-tage of not blocking the expression of conditioned responses

and thus allows any effects on extinction to be observeddirectly during the infusion session. Six rabbits were againtrained using the dual delay/trace procedure to asymptoticlevels of responding to both trial types. Each rabbit was thengiven two test sessions, separated by a day of retraining, to testthe effects of gabazine or ACSF on extinction. During thesetest sessions, gabazine (20 �M) or ACSF was infused into theinferior olive beginning 20 min before the start of extinctiontraining. As before, delay and trace trials were analyzed sep-arately with effects on delay responses serving as an indicationthat extinction in the cerebellum was blocked by the infusion.

Similar to the effects of muscimol infusions in the DCN,gabazine infusions in the inferior olive prevented the extinctionof delay but not trace eyelid responses. For delay conditioning,the average response likelihood during the gabazine infusionwas greater than during the ACSF infusion session [Fig. 5, A,C, left, and D, left; F(1,5) � 64.20, P � 0.001; sessions effect].

Fig. 4. Inactivation of DCN prevents the extinction of delay but not trace eyelid responses. A: response likelihood for delay (left) and trace conditioning trials(right) is plotted as a function of training block for each session type. *P � 0.05; **P � 0.01. Data from the postmuscimol session (dark gray) are shown twicefor comparison purposes. For delay conditioning, postmuscimol responding was similar to control 1 responding, indicating that extinction during the infusionwas prevented. For trace conditioning, the postmuscimol responding was similar to control 2 responding, indicating that extinction during the infusion wasunaffected. B: difference scores, obtained by subtracting performance during postmuscimol from either that during either control 1 or control 2, are plotted asa function of training block. Error bars represent SE. C: point-by-point averages of eyelid position sweeps of 8 subjects during postmuscimol, control 1, andcontrol 2 sessions.




Dow

nloaded from


Furthermore, response likelihood during the gabazine infusiondid not decrease across blocks (P � 0.90; blocks effect),whereas it did during the ACSF infusion [F(11, 55) � 17.63,P � 0.001; blocks effect]. For trace conditioning, although theaverage response likelihood during the gabazine infusion washigher than during the ACSF infusion [F(1, 5) � 7.44, P �0.04], responses during both sessions extinguished [F(11,55) �7.53, P � 0.001; blocks effect] at similar rates (P � 0.99;blocks � session interaction; Fig. 5, B, C, right, and D, right).Thus, although the gabazine infusion may have affected theextinction of trace eyelid responses, trace eyelid responsesnonetheless extinguished. Together with the DCN inactivationexperiments, these data suggest that the extinction of traceeyelid responses can occur via plasticity at a site outside of thecerebellum.

In an attempt to better understand the partial effect on theextinction of trace responses, examination of individual sub-jects’ performance revealed a bimodal effect of the gabazineinfusions (Fig. 6). The partial effect shown in Fig. 5 arises fromthe combined effects on four rabbits in which gabazine hadlittle to no effect on extinction and two rabbits in whichextinction was blocked. For all six of these rabbits, extinctionof delay responses was blocked, indicating that the infusionwas effective. The eyelid position sweeps in Fig. 6 are from

four of these six rabbits and demonstrate the range of effects ofthe gabazine infusion. In addition, the average response like-lihood for each rabbit during both the gabazine and ACSFinfusion sessions is shown in Fig. 6. Differences in cannulaplacements do not appear to account for these differences,because cannula were similarly placed in the anterior portion ofthe inferior olive (Fig. 6).

This rabbit-to-rabbit variability in the ability of the gabazineinfusion to prevent trace extinction, in the face of the ability ofthe same infusions to consistently prevent delay extinction,suggests that extinction of trace eyelid responses can be me-diated by plasticity at more than one site. One site requiresinhibition of climbing fiber inputs, similar to the mechanismthat mediates extinction of delay eyelid responses, and theother mechanism appears to involve plasticity outside of thecerebellum. On the basis of previous work demonstrating thattrace eyelid conditioning requires an input to the cerebellumdriven by the prefrontal cortex (Kalmbach et al. 2009) (Fig. 1),we hypothesize that extinction can be mediated by plasticitythat removes or shortens the duration of this input. Conversely,when this mechanism fails or occurs relatively slowly, traceextinction can occur by the cerebellar mechanism that involvesinhibition of climbing fiber inputs. As described below, wetested two key corollaries of this hypothesis: that cerebellarmechanisms are sufficient to mediate extinction of trace eyelidresponses (when the noncerebellar mechanism is bypassed)and that there is a bimodal distribution in the rate of traceresponse extinction, consistent with the potential contributionsof two different extinction mechanisms.

Cerebellar mechanisms can support extinction of trace eye-lid responses. Testing the hypothesis that cerebellar mecha-nisms are sufficient for extinction of trace eyelid responsesrequires a way to test extinction when the noncerebellar mech-anism is bypassed. We employed two methods to accomplishthis. The first is based on evidence that blocking the ability ofthe US to activate climbing fibers (Fig. 1C) induces cerebel-lum-mediated extinction, even during paired presentations ofthe CS and US (McCormick et al. 1985; Medina et al. 2002a).Thus, by pharmacologically blocking the US input to thecerebellum during paired training, we can induce cerebellum-mediated extinction while presumably any noncerebellarmechanism is still exposed to the CS and US in each trial. Twosubjects that were previously infused with gabazine in theinferior olive were retested using infusions of the AMPAreceptor antagonist NBQX during paired presentations of theCS and US. This infusion blocks the US input to the cerebel-lum, which has been shown previously to induce extinction ofdelay responses even when the CS is paired with the US(Medina et al. 2002a) but presumably does not block US inputto other brain regions. It is important with these infusions todistinguish between extinction and abolition via delayed dif-fusion to the essential region (Medina et al. 2002a; Zbarska etal. 2007, 2008). To do so, we began the infusion 40 min beforethe start of the session in one subject and 20 min before thesession in the other subject. Since extinction requires fewerthan 20 min, staggering the infusions in this way precludes thepossibility that diffusion over 20 min abolished the conditionedresponses (Medina et al. 2002a). As shown in Fig. 7, in bothsubjects the infusions induced extinction of delay [F(1,11) �17.127, P � 0.001; blocks effect] and trace eyelid responses[F(1,11) � 24.72, P � 0.001; blocks effect] at rates similar to

Fig. 5. Blocking the inhibition of climbing fibers prevents the extinction ofdelay but not trace conditioned responses. Response likelihood during delay(A) and trace conditioning trials (B) is plotted as a function of training block.Point-by-point average eyelid position sweeps of 6 subjects for ACSF (C) andgabazine infusion sessions (D) are shown below the graphs.




Dow

nloaded from


those seen during ACSF infusion (P � 0.73 for delay; P �0.33 for trace; blocks � session interaction). These data showthat the cerebellar mechanisms known to operate for delayresponses can support the extinction of trace responses undercircumstances where noncerebellar mechanisms are not enga-ged.

We also used a second method to test whether cerebellarmechanisms are sufficient for extinction of trace responseswhen noncerebellar mechanisms are bypassed. This test in-volved presenting inputs that roughly match the duration oftone and PFC-driven inputs to cerebellum that occur during traceconditioning via electrical stimulation of mossy fibers throughtwo separate electrodes (see Kalmbach et al. 2010a). Thepatterns of stimulation were designed to mimic the temporalduration of mossy fiber inputs to the cerebellum present duringtrace eyelid conditioning. Two stimulating electrodes wereimplanted in the middle cerebellar peduncle, where mossyfibers enter the cerebellum. A 500-ms train of pulses wasdelivered through one electrode to mimic tone-driven mossyfiber inputs, and a 1,000-ms train of pulses was delivered to thesecond electrode to mimic mossy fiber inputs driven by per-

sistent activity neurons in mPFC (Aitkin and Boyd 1978; Boydand Aitkin 1976; Campolattaro et al. 2011; Freeman andMuckler 2003; Siegel et al. 2009; Takehara-Nishiuchi andMcNaughton 2008). We have found previously that responsesacquired using these patterns of mossy fiber stimulation quan-titatively reproduce the properties of trace conditioned re-sponses (Kalmbach et al. 2010a). In this study we testedwhether extinction of these trace-like responses is preventedwhen cerebellar-mediated extinction mechanisms are blockedvia infusions of muscimol into the DCN. The experimentaldesign was identical to that employed in the first experiment(e.g., Figs. 3 and 4).

In five subjects, the effects of inactivating the DCN onextinction were assessed using the same test sessions as pre-sented in Fig. 2 following training with mossy fiber stimulationas the CS (example electrode placements are shown inFig. 8E). As before, infusing muscimol into the DCN abolishedthe expression of conditioned responses [Fig. 8; 23.6 � 3.9%responding during ACSF infusion, 2.2 � 2.2% respondingduring muscimol infusion; t(4) � 4.13, P � 0.004; pairedt-test]. However, this time the muscimol infusion prevented the

Fig. 6. Blocking climbing fiber inhibition prevents the extinc-tion of trace conditioned responses in some subjects. Eyelidposition sweeps during the ACSF and gabazine infusionssessions for 4 subjects demonstrate the range of effects. Thescatter plot at bottom left shows trace response likelihoodplotted as a function of delay response likelihood for eachsubject during the ACSF (light gray) and gabazine (black)extinction sessions. Symbols correspond to the behavioralsweeps shown above and cannula placements shown at right.Cannula placements are ordered from most posterior (top left)to most anterior (bottom right). Whereas trace extinction wasprevented by gabazine infusions in 2 subjects (e.g., triangleand circle), trace responses extinguished in the remaining 4subjects.




Dow

nloaded from


extinction of stimulation-mediated trace-like responses. Theaverage response likelihood during the entire postmuscimolsession and the control 1 session was not different (P � 0.14;paired sample t-test), but both were different from responselikelihood during the control 2 session [Fig. 8, B and C; t(5) �4.09, P � 0.01 for postmuscimol; t(5) � 6.25, P � 0.003 forcontrol 1; paired t-tests]. Furthermore, the rates at whichresponses extinguished during control 1 and postmuscimolsessions were not different (compare light gray and dark graycurves in Fig. 8B; P � 0.67; blocks � session interaction).Figure 8 also shows the difference scores, which demonstratethat response likelihood during the control 1 and postmuscimolsessions did not differ across training blocks, whereas responselikelihood during the control 2 session was less than thatduring postmuscimol sessions for the first couple of blocks.Together, these data provide evidence that the previouslyidentified cerebellar mechanisms of extinction of delay re-sponses can also mediate extinction of trace-like eyelid re-sponses when noncerebellar mechanisms are bypassed by theprocedures of this experiment. Thus extinction of trace eyelid

responses can be mediated by different sites, one in the cere-bellum and one outside of the cerebellum.

The rate of extinction for trace eyelid responses shows abimodal distribution. Finally, we tested a second corollary ofthe hypothesis, stating that two different mechanisms can, indifferent subjects or under different circumstances, mediateextinction of trace responses. If, as hinted by the above-described data (Fig. 6), these mechanisms impose somewhatdifferent rates of extinction, then we should expect a bimodaldistribution of extinction rates to be apparent in a sufficientlylarge sample of data. To test this corollary, we examined thenumber of conditioned responses and the trials to criterionextinction (the first instance of 8 non-conditioned responses in9 consecutive trials) in 21 trace-conditioned subjects. Asshown in Fig. 9, the distribution of both parameters differedfrom an expected normal distribution (black curve, Fig. 9A),with the majority of subjects extinguishing within the firstcouple of blocks and a minority extinguishing more slowly,nearly halfway through the session (P � 0.01 for both trials tocriterion and number of conditioned responses; Kolmogorov-Smirnov test for normality). This apparent bimodality can beseen most clearly by comparing the trials to reach criterionwith the number of conditioned responses during the extinctionsession for each subject (Fig. 9B). Also shown in Fig. 9 arebehavioral sweeps from a subject that extinguished quickly andanother that extinguished slowly. These data are consistentwith the idea that the extinction of trace conditioned responsescan be mediated by mechanisms operating in at least two sitesand that extinction mediated by one site is faster than extinc-tion mediated by the other site.

DISCUSSION

These results provide evidence that extinction of forebrain-dependent trace eyelid responses can occur via mechanismslocated in more than one brain region. We used a within-animal design to compare the effects of blocking a signal thatis necessary for extinction mechanisms in the cerebellum ondelay and trace responses (see Table 1 for a summary of theexperiments and key comparisons). Whether this signal wasblocked directly by infusions of gabazine in the inferior olive(Medina et al. 2002a) (Fig. 5) or indirectly by inactivation ofthe DCN (Hardiman et al. 1996; Ramnani and Yeo 1996) (Fig.4), we observed results consistent with previous findings: theextinction of delay responses was prevented. Our key finding isthat these manipulations do not necessarily prevent the extinc-tion of trace conditioned responses. Indeed, in most animalsgiven inferior olive infusions and in all animals given DCNinfusions, trace conditioned responses extinguished even whiledelay conditioned responses did not (Fig. 6). Apparent differ-ences between DCN and inferior olive manipulations on traceextinction may be due to differences in the sensitivity of eachtest. For the muscimol in the DCN infusions, effects onextinction were inferred indirectly based on animals respond-ing the next day, whereas the effects of gabazine into the olivecould be seen directly during the first day of extinction.Nevertheless, because trace extinction was relatively unaf-fected by manipulations of two distinct regions separated byseveral millimeters, it is likely that decreases in trace eyelidresponse were caused by extinction rather than the abolition ofresponses (e.g., by diffusion of the drug to a currently uniden-

Fig. 7. Infusion of 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(f)quinoxaline(NBQX) into the inferior olive causes extinction of delay and trace conditionedresponses. Response likelihood for delay (A) and trace conditioning trials (B)are shown for the gabazine, NBQX, and ACSF infusion sessions. Eyelidposition sweeps for the 2 subjects that received the NBQX infusion are shownin C. Symbols correspond to the data presented in Fig. 6. Notice that despitethe fact that the CS and US were paired, learned responses diminished at a ratesimilar to that during extinction training while ACSF was infused. Eyelidclosure in the lightest gray regions represents reflexive responses to the US.




Dow

nloaded from


tified region necessary for trace conditioning). Thus the datasupport the hypothesis that the extinction of delay conditionedresponses is signaled by the inhibition of climbing fibers bycollaterals in the DCN (Bengtsson et al., 2004; Hesslow andIvarsson 1996; Medina et al. 2002a; Svensson et al. 2006),whereas trace conditioned responses can undergo extinctionindependent of this process.

The observation that for some subjects the gabazine infu-sions in the inferior olive prevented trace extinction (Fig. 6)suggests that the cerebellar mechanisms can support extinctionof trace responses when, for whatever reasons, the noncerebel-lar mechanism is not engaged. In two separate experiments we

tested the hypothesis that the cerebellum mechanism is suffi-cient to mediate extinction of trace responses when the non-cerebellar mechanism is bypassed. First, we found that block-ing the US input to the cerebellum (Mauk et al. 1986; Mc-Cormick et al. 1985; Medina et al. 2002a) with infusions intothe inferior olive of the AMPA receptor antagonist NBQXcaused the extinction of delay and trace conditioned responses(Fig. 7), even though each CS was being paired with the US.Consistent with a previous report (Medina et al. 2002a), thiseffect was independent of when (20 or 40 min) the infusionwas delivered before the start of the session. Thus the decreasein responding after the infusion does not reflect the delayed

Fig. 8. Inactivation of DCN prevents the extinction ofresponses when trace conditioning is mimicked by usingelectrical stimulation of 2 sets of mossy fibers as the CS.A: response likelihood plotted as a function of trainingblock reveals that muscimol infusion in DCN preventedthe expression of stimulation-trained responses (P �0.001; paired sample t-test of response likelihood duringcontrol 1 and muscimol extinction session). Also shownare point-by-point averages of eyelid position sweeps of5 subjects during the muscimol infusion. B: responselikelihood during postmuscimol, control 1, and control 2extinction sessions is plotted as a function of trainingblock. *P � 0.05. C: point-by-point averages of eyelidposition sweeps of 5 subjects during each session.D: session difference scores plotted as a function oftraining block. *P � 0.05. E: coronal sections of thecerebellum of 2 subjects shows sample stimulating elec-trode (top) and cannula placements (bottom).




Dow

nloaded from


abolition of responses due to “cerebellar malfunction” as pre-viously suggested (Zbarska et al. 2007, 2008) but is insteaddependent on the number of training trials presented after theinfusion. Second, using direct stimulation of mossy fibers tomimic the tone and mPFC-driven input to the cerebellum thatoccur during trace conditioning (Aitkin and Boyd 1978; Boydand Aitkin 1976; Freeman and Muckler 2003; Kalmbach et al.,2010a; Takehara-Nishiuchi and McNaughton 2008), we pro-vide evidence that extinction can occur through cerebellarmechanisms under circumstances where the noncerebellar

mechanism cannot operate (Fig. 8). These observations alsosuggest that differences between delay and trace extinctionwith a tone CS are not due solely to differences in ISI (500 vs.1,000 ms).

These results suggest that a noncerebellar mechanism me-diates extinction of trace eyelid responses and that the previ-ously identified cerebellar mechanism is sufficient to extin-guish trace responses when the noncerebellar mechanism isbypassed or not engaged. This animal-to-animal variability inwhichever of the two mechanisms mediates extinction suggests

Fig. 9. The rate of trace conditioning extinction isbimodally distributed. A: cumulative frequency histo-gram of the trials to reach criterion extinction (top) andthe number of CRs during the first day of extinction for21 trace-conditioned subjects. The gray line representsthe sample data, and the black line represents the ex-pected normal distribution. B: the bimodal extinction canbe seen by comparing the trials to reach criterion withthe number of CRs for each subject (circles). The squarerepresents the mean value for all subjects. Gray symbolscorrespond to the eyelid position sweeps in C, whichshow examples of a subject who extinguished quickly(top) and another who extinguished slowly (bottom).

Table 1. Summary of experiments and key comparisons

Experiment (Fig.) Key Comparisons Behavioral Prediction

Muscimol (1 mM) or ACSF control in deepnucleus during extinction (Figs. 2–4)

Postmuscimol day 1 responding vs. control day 1(ACSF infusion) and control day 2 for bothdelay 500 and trace 500/500

If muscimol blocked extinction (delay or trace),then postmuscimol responding � control day1, both �control day 2

Gabazine (20 �M) in inferior olive duringextinction (Fig. 5–6)

Responding during extinction (gabazine vs.ACSF) for both delay 500 and trace 500/500

If gabazine blocked extinction, then respondingduring gabazine infusion � respondingduring ACSF infusion

NBQX (150 �M) during paired training (Fig. 7) NBQX during paired conditioning vs. ACSFduring extinction training for both delay 500and trace 500/500

If NBQX caused extinction, then respondingduring ACSF extinction � responding duringNBQX training

Muscimol (1 mM) in deep nucleus duringextinction with mossy fiber stimulation as theCS (Fig. 8)

Postmuscimol day 1 vs. control day 1 (ACSFinfusion) and control day 2

If extinction is blocked, then postmuscimol �control day 1, both �control day 2

Rate of extinction during trace conditioning for21 subjects trained with tone CS (Fig. 9)

Trace extinction: distribution of trials to criterionand number of CRs vs. predicted normaldistribution

Cumulative frequency of trials to criterion andnumber of CRs diverge from expectednormal distribution

ACSF, artificial cerebrospinal fluid; CR, conditioned response; CS, conditioned stimulus; NBQX, AMPA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(f)quinoxaline.




Dow

nloaded from


that the rate of extinction for trace responses should display abimodal distribution. Examining trace responses in 21 subjects,we found that the rate of extinction during the first day ofextinction training was indeed bimodally distributed; mostanimals extinguished quickly within the first couple of blocks,whereas a few subjects extinguished more slowly approxi-mately halfway through the session (Fig. 9). Although thesedata are consistent with the hypothesis that trace conditionedresponses can extinguish via more than one mechanism andthat these mechanisms produce different rates of extinction,other explanations are possible.

What is the noncerebellar mechanism of trace extinction?Although our experiments do not directly address the locationor mechanism of the noncerebellar mechanism, there are atleast two general possibilities: 1) descending brain systemscould inhibit the motor systems that control eyelid closure, or2) the mPFC-driven input to the cerebellum that is required forthe expression of trace conditioned responses could diminish orbe inhibited (Kalmbach et al. 2009).

In either case, the cerebellum-independent extinction pro-cess may occur in the forebrain. Whereas lesions or reversibleinactivation of the cerebellum affect both delay and traceeyelid conditioning (Kalmbach et al. 2009; McCormick andThompson 1984; Pakaprot et al. 2009; Woodruff-Pak et al.1985), only trace conditioning is affected by lesions of fore-brain structures such as the hippocampus, mPFC, and primarysensory cortex (Galvez et al. 2007; Kalmbach et al. 2009; Kimet al. 1995; Moyer et al. 1990; Powell et al. 2001; Solomon etal. 1986; Takehara et al. 2003; Weible et al. 2000). Thus it ispossible that plasticity in these sites can contribute to theextinction of trace conditioned responses. Indeed, lesions ofhippocampus, infralimbic area of mPFC, and other forebrainstructures have been reported to affect the extinction of traceconditioned responses with long trace intervals, short traceintervals, and even delay conditioned responses (McCormickand Thompson 1982; Moyer et al. 1990; Weible et al. 2000).Ongoing studies are investigating the extracerebellar sites in-volved in the extinction of trace conditioned responses.

The first possible extracerebellar extinction mechanism, thatdescending brain systems inhibit the motor systems that controleyelid closure, parallels evidence that the extinction of learnedfear is partly mediated by the inhibition of the brain systemsresponsible for the expression of fear memories (Maren andQuirk 2004; Milad and Quirk 2002; Milad et al. 2004; Quirkand Mueller 2008). In eyelid conditioning, learning in thecerebellum is expressed through its connections with the rednucleus, which in turn projects to the premotor and motorneurons that innervate the orbicularis oculi muscle responsiblefor eyelid closure (Morcuende et al. 2002; Thompson andSteinmetz 2009). Because these areas receive monosynapticand polysynaptic input from various forebrain regions (Ber-nays et al. 1988; Morcuende et al. 2002), the noncerebellarextinction mechanism could parallel the fear extinction path-ways by involving descending inhibition of these sites. Thepresent work and data from previous studies, however, imposea specific constraint on this mechanism: this downstreaminhibition would have to occur only for trace and not for delayconditioning. This constraint is imposed by observations thatmanipulations of cerebellum, but not red nucleus, prevent theextinction of delay conditioned responses (Figs. 3 and 4)

(Hardiman et al. 1996; Medina et al. 2002a; Ramnani and Yeo1996; Robleto and Thompson 2008).

The second possibility, that mPFC-driven input to the cere-bellum diminishes or is inhibited, is based on recent evidenceshowing that learning in the cerebellum during trace eyelidconditioning is in response to mPFC-driven input that persiststhrough the trace interval (Kalmbach et al. 2009, 2010a; Siegelet al. 2009). This is in contrast to delay conditioning, wherecerebellar learning occurs in response to mossy fiber inputs thatare more directly driven by the auditory system (Campolattaroet al. 2007; Freeman et al. 2007; Halverson et al. 2008;Halverson and Freeman 2010; Steinmetz et al. 1987). Thisdifference suggests that extinction of trace conditioned re-sponses could occur through a currently unidentified mecha-nism that results in the omission of mPFC-driven input to thecerebellum.

Although the plausibility of this hypothesis must be evalu-ated by future experiments, we propose the following concretebut speculative hypothesis regarding the mechanisms of ex-tinction for trace eyelid responses. During the acquisition oftrace eyelid conditioning, neurons in prefrontal cortex acqu-ire the ability to respond to the CS with activity that persiststhrough the trace interval to overlap with the US and in thisway engage cerebellar learning. This input is necessary for theexpression of trace eyelid responses as revealed by reversibleabolition of trace response expression by inactivation of mPFCwith infusion of the GABAA receptor agonist muscimol(Kalmbach et al. 2009). As such, the noncerebellar mechanismof extinction may involve changes in this forebrain-drivenresponse that render it unable to support expression of condi-tioned responses (either suppressing it or shortening its dura-tion). This hypothesis is consistent with the relatively abruptdisappearance of trace responses during extinction: once theforebrain-driven response is omitted or foreshortened, traceresponses would disappear immediately. To the cerebellum, itwould be the equivalent of omitting the CS. We furtherhypothesize that, for unknown reasons, this forebrain mecha-nism is sometimes not successful, in which case the cerebellarmechanism would be sufficient to mediate extinction of thetrace responses. This aspect of the hypothesis is consistent withthe bimodal distribution of extinction rates we present in Fig.9. It is also consistent with evidence that the extinction of traceeyelid responses can occur in the cerebellum when the requiredforebrain input is ensured (Figs. 7 and 8). Ongoing single-unitrecording studies are testing this hypothesis more directly.These studies should reveal neurons in mPFC and pontinenuclei that show tone-evoked persistent activity that extin-guishes at the same rate as behavioral extinction in someanimals and more slowly than behavioral extinction in others.

In summary, we have provided evidence that trace eyelidresponses can be extinguished via processes operating at morethan one site. This shows that when the expression of a learnedresponse requires plasticity in multiple brain systems, as is thecase for trace conditioning, not only can extinction of thatresponse be mediated by processes operating at more than onesite, but there may be subject-to-subject or situation-to-situa-tion differences in the relative contributions from those sites.Trace eyelid conditioning offers a unique opportunity to iden-tify the variables that determine the factors, sites, and mecha-nisms by which such responses extinguish.




Dow

nloaded from


ACKNOWLEDGMENTS

We thank Frank Riusech for assistance with histology and surgery.

GRANTS

This work was supported by National Institute of Mental Health Grants MH46904 and MH 57051.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

B.E.K. and M.D.M. conception and design of research; B.E.K. and M.D.M.performed experiments; B.E.K. and M.D.M. analyzed data; B.E.K. andM.D.M. interpreted results of experiments; B.E.K. and M.D.M. preparedfigures; B.E.K. and M.D.M. drafted manuscript; B.E.K. and M.D.M. editedand revised manuscript; B.E.K. and M.D.M. approved final version of manu-script.

REFERENCES

Aitkin LM, Boyd J. Acoustic input to the lateral pontine nuclei. Hear Res 1:67–77, 1978.

Bengtsson F, Svensson P, Hesslow G. Feedback control of Purkinje cellactivity by the cerebello-olivary pathway. Eur J Neurosci 20: 2999–3005,2004.

Bernays RL, Heeb L, Cuenod M, Streit P. Afferents to the rat red nucleusstudied by means of D-[3H]aspartate, [3H]choline, and non-selective tracers.Neuroscience 26: 601–619, 1988.

Boyd J, Aitkin L. Responses of single units in the pontine nuclei of the cat toacoustic stimulation. Neurosci Lett 3: 259–263, 1976.

Campolattaro MM, Halverson HE, Freeman JH. Medial auditory thalamicstimulation as a conditioned stimulus for eyeblink conditioning in rats.Learn Mem 14: 152–159, 2007.

Campolattaro MM, Kashef A, Lee I, Freeman JH. Neuronal correlates ofcross-modal transfer in the cerebellum and pontine nuclei. J Neurosci 31:4051–4062, 2011.

Chapman PF, Steinmetz JE, Sears LL, Thompson RF. Effects of lidocaineinjection in the interpositus nucleus, and red nucleus on conditioned behav-ioral and neuronal responses. Brain Res 537: 149–156, 1990.

Fontan-Lozano A, Troncoso J, Munera A, Carrion AM, Delgado-GarciaJM. Cholinergic septo-hippocampal innervation is required for trace eye-blink classical conditioning. Learn Mem 12: 557–563, 2005.

Freeman JH, Halverson HE, Hubbard EM. Inferior colliculus lesionsimpair eyeblink conditioning in rats. Learn Mem 14: 842–846, 2007.

Freeman JH Jr, Muckler AS. Developmental changes in eyeblink condition-ing and neuronal activity in the pontine nuclei. Learn Mem 10: 337–345,2003.

Galvez R, Weible AP, Disterhoft JF. Cortical barrel lesions impair whis-ker-CS trace eyeblink conditioning. Learn Mem 14: 94–100, 2007.

Garcia KS, Mauk MD. Pharmacological analysis of cerebellar contributionsto the timing, and expression of conditioned eyelid responses. Neurophar-macology 37: 471–480, 1998.

Halverson HE, Freeman JH. Medial auditory thalamic input to the lateralpontine nuclei is necessary for auditory eyeblink conditioning. NeurobiolLearn Mem 93: 92–98, 2010.

Halverson HE, Poremba A, Freeman JH. Medial auditory thalamus inacti-vation prevents acquisition, and retention of eyeblink conditioning. LearnMem 15: 532–538, 2008.

Hardiman MJ, Ramnani N, Yeo CH. Reversible inactivations of the cere-bellum with muscimol prevent the acquisition, and extinction of conditionednictitating membrane responses in the rabbit. Exp Brain Res 110: 235–247,1996.

Hesslow G, Ivarsson M. Inhibition of the inferior olive during conditionedresponses in the decerebrate ferret. Exp Brain Res 110: 36–46, 1996.

Jirenhed DA, Bengtsson F, Hesslow G. Acquisition, extinction, and reacqui-sition of a cerebellar cortical memory trace. J Neurosci 27: 2493–2502,2007.

Kalmbach BE, Ohyama T, Kreider JC, Riusech F, Mauk MD. Interactionsbetween prefrontal cortex, and cerebellum revealed by trace eyelid condi-tioning. Learn Mem 16: 86–95, 2009.

Kalmbach BE, Ohyama T, Mauk MD. Temporal patterns of inputs tocerebellum necessary, and sufficient for trace eyelid conditioning. J Neuro-physiol 104: 627–640, 2010a.

Kalmbach BE, Davis T, Ohyama T, Riusech F, Nores WL, Mauk MD.Cerebellar cortex contributions to the expression, and timing of conditionedeyelid responses. J Neurophysiol 103: 2039–2049, 2010b.

Kehoe EJ. Repeated acquisitions, and extinctions in classical conditioning ofthe rabbit nictitating membrane response. Learn Mem 13: 366–375, 2006.

Kim JJ, Clark RE, Thompson RF. Hippocampectomy impairs the memoryof recently, but not remotely, acquired trace eyeblink conditioned responses.Behav Neurosci 109: 195–203, 1995.

Maren S, Quirk GJ. Neuronal signalling of fear memory. Nat Rev Neurosci5: 844–852, 2004.

Mauk MD, Steinmetz JE, Thompson RF. Classical conditioning usingstimulation of the inferior olive as the unconditioned stimulus. Proc NatlAcad Sci USA 83: 5349–5353, 1986.

Mauk MD, Thompson RF. Retention of classically conditioned eyelid re-sponses following acute decerebration. Brain Res 403: 89–95, 1987.

McCormick DA, Thompson RF. Locus coeruleus lesions, and resistance toextinction of a classically conditioned response: involvement of the neocor-tex and hippocampus. Brain Res 245: 239–249, 1982.

McCormick DA, Thompson RF. Cerebellum: essential involvement in theclassically conditioned eyelid response. Science 223: 296–299, 1984.

McCormick DA, Steinmetz JE, Thompson RF. Lesions of the inferiorolivary complex cause extinction of the classically conditioned eyeblinkresponse. Brain Res 359: 120–130, 1985.

Medina JF, Garcia KS, Mauk MD. A mechanism for savings in thecerebellum. J Neurosci 21: 4081–4089, 2001.

Medina JF, Nores WL, Mauk MD. Inhibition of climbing fibres is a signalfor the extinction of conditioned eyelid responses. Nature 416: 330–333,2002a.

Medina JF, Repa JC, Mauk MD, LeDoux JE. Parallels between cerebel-lum-, and amygdala-dependent conditioning. Nat Rev Neurosci 3: 122–131,2002b.

Milad MR, Quirk GJ. Neurons in medial prefrontal cortex signal memory forfear extinction. Nature 420: 70–74, 2002.

Milad MR, Vidal-Gonzalez I, Quirk GJ. Electrical stimulation of medialprefrontal cortex reduces conditioned fear in a temporally specific manner.Behav Neurosci 118: 389–394, 2004.

Morcuende S, Delgado-Garcia JM, Ugolini G. Neuronal premotor networksinvolved in eyelid responses: retrograde transneuronal tracing with rabiesvirus from the orbicularis oculi muscle in the rat. J Neurosci 22: 8808–8818, 2002.

Moyer JR Jr, Deyo RA, Disterhoft JF. Hippocampectomy disrupts traceeye-blink conditioning in rabbits. Behav Neurosci 104: 243–252, 1990.

Oakley DA, Russell IS. Neocortical lesions and Pavlovian conditioning.Physiol Behav 8: 915–926, 1972.

Ohyama T, Mauk M. Latent acquisition of timed responses in cerebellarcortex. J Neurosci 21: 682–690, 2001.

Ohyama T, Nores WL, Medina JF, Riusech FA, Mauk MD. Learning-induced plasticity in deep cerebellar nucleus. J Neurosci 26: 12656–12663,2006.

Pakaprot N, Kim S, Thompson RF. The role of the cerebellar interpositusnucleus in short and long term memory for trace eyeblink conditioning.Behav Neurosci 123: 54–61, 2009.

Perrett SP, Mauk MD. Extinction of conditioned eyelid responses requiresthe anterior lobe of cerebellar cortex. J Neurosci 15: 2074–2080, 1995.

Perrett SP, Ruiz BP, Mauk MD. Cerebellar cortex lesions disrupt learning-dependent timing of conditioned eyelid responses. J Neurosci 13: 1708–1718, 1993.

Peters J, Kalivas PW, Quirk GJ. Extinction circuits for fear and addictionoverlap in prefrontal cortex. Learn Mem 16: 279–288, 2009.

Phelps EA, LeDoux JE. Contributions of the amygdala to emotion process-ing: from animal models to human behavior. Neuron 48: 175–187, 2005.

Powell DA, Churchwell J. Mediodorsal thalamic lesions impair trace eye-blink conditioning in the rabbit. Learn Mem 9: 10–17, 2002.

Powell DA, Skaggs H, Churchwell J, McLaughlin J. Posttraining lesions ofthe medial prefrontal cortex impair performance of Pavlovian eyeblinkconditioning but have no effect on concomitant heart rate changes in rabbits(Oryctolagus cuniculus). Behav Neurosci 115: 1029–1038, 2001.

Quirk GJ, Mueller D. Neural mechanisms of extinction learning and retrieval.Neuropsychopharmacology 33: 56–72, 2008.




Dow

nloaded from


Ramnani N, Yeo CH. Reversible inactivations of the cerebellum prevent theextinction of conditioned nictitating membrane responses in rabbits. JPhysiol 495: 159–168, 1996.

Robleto K, Thompson RF. Extinction of a classically conditioned response:red nucleus and interpositus. J Neurosci 28: 2651–2658, 2008.

Rosen DJ, Steinmetz JE, Thompson RF. Classical discrimination condition-ing of the rabbit’s eyelid response using pontine stimulation as a conditionedstimulus. Behav Neural Biol 52: 51–62, 1989.

Ryou JW, Cho SY, Kim HT. Lesions of the entorhinal cortex impairacquisition of hippocampal-dependent trace conditioning. Neurobiol LearnMem 75: 121–127, 2001.

Siegel JJ, Kalmbach BE, Chitwood RA, Dembrow N, Johnston D, MaukMD. Prefrontal cortex and lateral pontine neurons display tone-evokedpersistent activity during trace eyelid conditioning. Abstr Soc Neurosci384.9, 2009.

Solomon PR, Vander Schaaf ER, Thompson RF, Weisz DJ. Hippocampus,and trace conditioning of the rabbit’s classically conditioned nictitatingmembrane response. Behav Neurosci 100: 729–744, 1986.

Steinmetz JE, Logan CG, Rosen DJ, Thompson JK, Lavond DG, Thomp-son RF. Initial localization of the acoustic conditioned stimulus projectionsystem to the cerebellum essential for classical eyelid conditioning. ProcNatl Acad Sci USA 84: 3531–3535, 1987.

Svensson P, Bengtsson F, Hesslow G. Cerebellar inhibition of inferiorolivary transmission in the decerebrate ferret. Exp Brain Res 168: 241–253, 2006.

Takehara K, Kawahara S, Kirino Y. Time-dependent reorganization of thebrain components underlying memory retention in trace eyeblink condition-ing. J Neurosci 23: 9897–9905, 2003.

Takehara-Nishiuchi K, McNaughton BL. Spontaneous changes of neocortical codefor associative memory during consolidation. Science 322: 960–963, 2008.

Thompson RF, Steinmetz JE. The role of the cerebellum in classical conditioningof discrete behavioral responses. Neuroscience 162: 732–755, 2009.

Weible AP, McEchron MD, Disterhoft JF. Cortical involvement in acqui-sition, and extinction of trace eyeblink conditioning. Behav Neurosci 114:1058–1067, 2000.

Weible AP, Weiss C, Disterhoft JF. Connections of the caudal anteriorcingulate cortex in rabbit: neural circuitry participating in the acquisition oftrace eyeblink conditioning. Neuroscience 145: 288–302, 2007.

Woodruff-Pak DS, Lavond DG, Thompson RF. Trace conditioning: abol-ished by cerebellar nuclear lesions but not lateral cerebellar cortex aspira-tions. Brain Res 348: 249–260, 1985.

Yeo CH, Hardiman MJ, Glickstein M. Classical conditioning of the nicti-tating membrane response of the rabbit. IV. Lesions of the inferior olive.Exp Brain Res 63: 81–92, 1986.

Zbarska S, Bloedel JR, Bracha V. Cerebellar dysfunction explains theextinction-like abolition of conditioned eyeblinks after NBQX injections inthe inferior olive. J Neurosci 28: 10–20, 2008.

Zbarska S, Holland EA, Bloedel JR, Bracha V. Inferior olivary inactivationabolishes conditioned eyeblinks: extinction or cerebellar malfunction? Be-hav Brain Res 178: 128–138, 2007.




Dow

nloaded from


multiple sites of extinction for a single learned...

Documents