.lounrul ol rhe lntarnutionul i\taux4ts.rr'hologicul Stx iclr' (2004), 10. 647 654. Copyright c 2004 INS. Pubiished by Cambridge Unir'crsiry Press. Prinled in the USA DOI: 10. I 0l 7/S I l-556 I 7704 I 0501 8 Procedural and declarative memory in obsessive-compulsive disorder ROBI,RT M. ROTH.I.J JACIN'IHE tsARItsI,AU.2,1 DENISE MILOVAN.s.{ KIERON O'CONNOR.I rx» CRISTO TODOROV' rNeuropsychology PrÙgranr- I)!-partment ofl'sychiatry. Dartmouth Medical School, Lebanon, NH :Laboratory oi Human Neuropsychology and Neurrrphysiology, Déparlemcnl de Psychologre, Unirersiti' Laval, Qucbec, Canada rl)cparlnrcnl trf Psvchologv, C'oncordia Unircrsity'. \{ontrcal, Qucbec. Canada rcentre dc Rccherche Fcmand Seguin. Hopital Louis-H. Lalbntainc, Montreal. Quebec. Canada 5Hopil;tl Louis-H. Lalirnlaine, Montl!'â1. Quebcc" C'anada (Rri( FrvFr) May 8. 20()-1; Rr:r'rsr.:o Oclober 29. 200]: Ac(-Êrpr Ft) Fcbruary 2, 2004) Âbstract Obsessivc-compulsivc disorder (OCD) has been associate<i with lrontostriatal abnormality. This has led to the hypothesis that thc disordc-r is characterizcd by abnorrnality ofprocedural nrenrory. However, evidencc for either procedural ordeclarative lrcrnory disturbance has been mixed, and ferv studies have dircctly assessed both ofthesc fbnns ol'r.nenror-v- in thc sanre patient group. In thc present study. we assessed encoding and retrieval in declalative r.rlcr.nory using the Rey Auditory Verbal Leaming Tesr ( RAVLT). and procedural Inelnory using the Pursuit Rotor Task. in 27 adults q,ith OCD and 29 nralclied hcalthy controls. Groups did not dilfèr with respcct to salient dernographic characteristics or nlenlory on lhe R,AVLT. ln contrast. patients with OCD pcrlbrmcd significantly brlter than controls during thc early^ but not latcr trial blocks of the Pursuit Rotor Task. This pattern ol results indicatcs intact encoding and retricval in declalative meurory. but abnorrnally enhanced procedural rnemory during the early course of'learning in OCD. These findings uray be consistent rvith striatal overactivalion observed in neuroinraging studics ofOCD, as well as lhe p«rnrinent role ofthe striatuur during early stages ofproceclural lrrenrory. (J/^§. 2004. 10.611 654.) Kcywords: OC'D. Procedural nreurory. Declarative nlcnlory. Striaturn OCD has been associated witlr tiont«:striatal circuitry abnonnality. Positron ernissioll tolllography ( PET) studies have reported abnormally increased blood f'low in several brain regions while patients were rcsting or during symp- INTRODUCTION Obsessive-compulsive disorder (OCD) is a neuropsychiat- ric condition charactelized by intrusive, repetitive. and unrvantcd thoughts that usually prornpt repetitive overt tom provocation ( Rauch et al.. 200 I I Saxena & Rauch. 2000). These r-egions have generally included the striatum (especially the caudate nucleus). orbitofrontal gyrus and other fiontal lobe regions. thalarnus, and cingulate gyrus. Tteatment with selective serotonin reuptake inhibitors or behavioral therapy has been found to produce a nonnalization of regional ovemctivation (Saxcna et al., 2001, 2002). Ciornputerized tornography and magnetic resorrance irnaging studies have reported structural abnormalities in striatum and orbitofrontal gyrus. although inconsistently (Aylward et al., i996; Robinson ct al.. 1995; Rosenbcrg et al., 1997; Szcszko ancl/or co'r,crt behaviors intcndcd to rcduce anxiety associated with the thoughts. Lifetirnc prcvalence rates for OCI) range liom approximately one to three percent in the adult population (tlebbinglon, 1998; Honvath & Weissrnan, 2000). Thc disorder generally has its onset in childhood or adolescence, and is characterized by a chronic. rvaxing and waning course (Skoog & Skoog. 1999). Signiticantly reduced quality of lit'c, as well as social, f-tunily and occupational problerns are cornmon in OCD (Amir et a[.. 2000; Bystritsky et al.. 2001). Rcprinl rctlucrls to: Rohcrt et i, *,,t', *.n. n.,'.n".n","*t al., 1999). The presence of frontostriatal ciLcuitry abnorrnality sug- Progranr. Departnrent of Psychiatry. Dartntouth \'lcdical School, One \lcdical Cenler [)r'irc. [-ebanon. NH 0.1756-000 l. [-nrail: Robcrt.lv{.Rothiri gests that patieuts with OCD nray dcrnonstrate pertbrmauce dcficits on procedural melllory tasks. Procedural Darlnrtrrrth.ed u. 647 R.M. Roth et al. 648 lnclnory may be conccptualized as leanling without awarellcss and can be obserrvcd in the irnprovernent ofspeed and/or accuracy ou tasks such as implicit seqtlerlce learning. rnirror rcading, probabilistic classitlcation, star mirror tracing. or the Pursuit Rotor Task (Squire. 1986). lt should be noted that procedural rnemory likely involvc's several processes, as dernonstratcd by dissociations in perfortnance on dillèrcnt proccdural rncmory tasks u,ithin thc samc group of 'l patiellts (Cabricli et al.. c)97). Nonethcless. uumerous studies using a variety oftasks have shorvn that aspects ofproccdural rncrrory are abnorrnal in patients rvith basal ganglia dyslirnctiorr such as thosc rvith Tourette's syndrotne, Parkinson's alld Huntington's diseasc. as lvcll as anitnals rvith damage- to tl'ontostriatal circuitry ( Bylsma et al.. 19901 Gorncz Bcldarrain et al.. 1999; Hcindcl ct al., 19881 Miyachi ct al.. 2002: Packard & Krrorvlton, 2002; Pascual-Leone et al., 1995; Sarazin et aI.,2002; Stebbins et al., 1995). Il1 contrast to procedural lnelnoly, cleclarative memory involves thc couscious lcarning and lccall of infbrnration such as Iists olrvords or images (Gabricli, 1998). Dysfurrction in the encoding and retrieval of declartrtive tnetnot'ies has bcen most coulmorrly associated rvith darnagc to mesial tcrnporal lobe structures, particularly thc hippocampus (Cabeza & Nyberg, 2000: SqLrire & Zola. 1996). Double dissociations may bc observed rvhereby patients rvith basal ganglia abnonnality pcrtbrrn poorly on procedural memory tasks and relativcly rvcll in their encoding and rctriel'al of declalative rrL'lnories, rvhilc patier-rts with mesial ternporal lobe abnonnalitics sholv thc revcrse pattern (Gabrieli, 1 9981 Sabc ct al., 1995). To datc. no consistcnt cvidencc fbrnresial temporal lobc abnomrality has beerr tbund in OCD. It has been hypothesized that because of a procedural rnenlory dcficit, infonnation that would othcrwisc be processed without awarelress rnust be plocessetl with full awareness bv patients with OCD. resulting in incleased arousal and misallocation of attcntion ( Rauch & Savage. 2000: Rauch ct al.. t9!)7). Dcclaratil'c rlemory circuitry is recruited in order to colrlpcnsatc for abrtonnal procedr.rral melnory. albeil somovhal inelliciently. F unclional ncuroimaging studics using procedural mcrnory tasks have observed abnormal recruitrncnt of l'r'ontostriatal circuitry and increascd lrippocampal acti!'atioû relative to hc'althy controls. supporting this hypothesis (Rauch et al., 1997: Rotl.r et al..2003). Fcrv bchavioral studics have assesscd proccdural mcmoly in OCD, and only a subset has cvaluated both procedural and declarative mernory iu the same group of patients. Patienfs with OCD rvcre shorvlt to perfbnn more poorly than hcalthy controls on thc proccdural tnctnory cotnponeltt of the Torver of Hanoi (Cavedini et al.. 2001). as rvell as on an irnplicit sequence learning task rvhere the opportunity to recluit declarative rncmoty rvas intertèr'ed rvith through dual task dcmands ( Dcckersbach et al.. 2002). I11 corltrast. olle study reported poor delayed recall fbr a rvord list but intact Pursuit Rotor Task pcrfbrurance ( Martin et al.. 1993). Other invcstigations of declarative rnclnory in OCD have yiclded inconsistent lin<lings (Christensen ct al., 1992: ['lollander et al.. 199(): Savage et al.. 1996). When dellcits iu eucodin-e and retrieval in declarative mcmory are observed in OCD, evidence suggests that thcy arc largely due to in-rpaircd usc of strategic processes subserved by frontostriatal circuitry rathcr tharl processes subserved by rncsial ternporal lobe circuitry (Deckelsbach et al., 2000; Savage et al., 1999, 2000). The conflicting rcsults reported rvith respect to proce«lural rncmory in OCD rnay be partly due to thc usc of relatively small sample sizes, the inclusion of medicated patients, and tàilure to exclude patients rvith comorbid rnajor deplessive disordcr. With regard to thc latter possibility. patients rvith prirnary rnajor depressive disorder have bcen rcportcd to pcrfonx poorly on procedural memory tasks (Sabe et al.. 1995), and even non-clinical levels ofdepression have becn associatcd with reduced procedural lnemory (Kalechstein et al., 1998). In the prcsent study, rve sought to evaluate the hypothesis that procedural memory is abnormal in a sample of unmedicated and non-clinically depressed paticnts with OCD. We hypothcsized that, relative to healthy controls. the patient group rvould show poorer performance on the Pursuit Rotor Task. a lneasure of procedural rnemory. and norrnal pcrforrnance on thc Rey Auditory Verbal Learr-ring Test, a task designed to assess encoding and retrieval in declarative lnelnol'y. METHODS Research Participants Participants rvere 27 adult outpatients with OCD and 29 healthy controls. All participants were right-handed by selfrcportcd hand preferencc for rvriting, and u'ere betwcen thc ages of I 8 and 65 years. None of the participants had a history olsubstâuce abuse. neurolo-qical illness, head injury resulting in a loss ofconsciousness, elcctroconvulsive therapy or psychosurgcry, or othcr mcdical illncss r,r,ith potcntial effects on cognitivc tinctioning. Paticnts met diagrlostic criteria tbr OCD as detlned by the Diagnostic and Statisti- cal Manual of Mental Disorders-Fourth Edition criteria (Aurcrican Psychiatric Association, 1994). Diagnosis was establishcd using the Anxiety Disorders Intervierv Schedule (ADIS). a structurcd diagnostic interview for anxiety disorders and exclusionary conditions bascd on DSM-lV criteria (l3rown et al.. 1994). The ADIS rvas administered by either a psychiatrist or psychologist with extensivL- expericncc in OCD. Patie nts rvere excludcd if they had a comorbid DSM-lV diagnosis of rnajor deprcssivc disorder or dysthyrnia. Arnong the patients rvitlr OC D.4 mct diagnostic criteria lbr comorbid generalized anxiety disorder. -l had social phobia. and 1 had parlic disorder. However. in all câscs OCD was clcarly the prirnary diagnosis. OCD syrnptorn severity was assessed using the Yale-Brorvu ObsessiveCornpulsive Scale (YBOCS: Goodrnan et al.. 1989). The YBOCS is a structurcd clinical intervierv specilÏcally designed to detennine the naturc. extent and severity of obscssions and compulsions. Ten items pertaining to the Metnor.t'in OCI) 649 severity of obsessions and compulsions arc rated or1 a fbur point scale of seve'r'ity, u,ith higher scores reflecting greater severity of illness. Arnong our patienls with OCD. l9 were ncver prescribed psychotropic mcdications and 8 had discoutinued thc usc ofselectivc serotonin reuptake inhibitors lbr a minirnurn of trvo weeks prior to their participation in the study. None of the patients werc undergoing cognitivebchavioral therapy at thc tirnc of participation. Hc'althy controls were recruited frorn a subject pool at the Centre de Recherche Fernand Seguin (a dedicated research center affil- iated with the University of Montreal and Hopital Louis H. Lafbntaine) and advertiscrnents in a local nervspapcr, antl Irad no irersonal lristory ofpsychiatric illncss based on self- report. All participants plovided written inlbrmed conserlt al'ter cornplete dcscription of the study. spced during baseline at which they were able to track the targct closest to au averagc of5 s was used exclusively for the test trials. This procedure was ernployed to minirnize absolute group dillerence in motor coordination or spced, and is consistent with prior studies using this task (van Gorp et al.. 1999). Participants next completed a total of six test blocks of four twenty second trials each. The inter-trial resting pcriod was 20 s, while the inter-block resting pcriod was 60 scconds. Thirty nrinutes of other cognitive tests (results to be reported elsewhere) were completed betrveen Blocks two and three. and again between Blocks 4 and 5. The average pcrcent time on target fbr each block was calculated by averaging tlre tirne ou larget for all four trials within a block. Data Analysis Procedure Group differences in demographic and clinical characteristics were analyzed using independent samples r tests and Mood Thc Beck Depression Inventory (BDl) is a 21-item sell'report scalc designed to assess ibr cognitive, emotional and somalic syrnptonls ofdepression ( Bcck. 1 987). Scores range liom 0 to 63. with higher scores indicating a greater level of deprcssion. Situational anxiety u,as rneasured u,ith thc Spiclbcrger Statc-Trait Anxicty Inventory. Statc subscale (STAI-S; Spielberger ct al.. 1970). Scores ou the STAI-S range fion.r 20 to U0, rvith higher scores indicating a greâter degree of anxiety. nonparametric statistics, as appropriate. Group X Trial and Group X Block rnixed model repeated measures analysis of variance were conducted for the RAVLT learning trials and Pursuit RotorTask, rcspectively. Indcpendcnt sarnple l-tests rvere carried out to evaluatc group dilferences on the iudividual trials of the RAVLT and blocks of the Pursuit Rotor Task. All tests uscd two-tailed comparisons with signiticance lcvcl set at p < .05. RESULTS Declarative nrcnlonl On the Rey Auditory Verbal Learning Test (RAVLT; Spreen & Strauss. 19981, participarlts were rcad aloud a list of flfteen rvords and aske«I to recall as many as possible. The procedure was repeated fbr a total ol five leaming trials. Irollorving thc trtth leaming trial. an interf'erencc list of flfieerr nerv rvords rvas preseuted followed by frce recall of the new list. Participants wcre subsequently asked to recall the rvords tiorn the flrst list (short dclay). Subjects were asked to recall thc llrst list a-eain afler a 30-min dclay (long delay), arrd then complcted a yes/no rccognition menlory trial. Total wolds recalled. as rvell as pcrcent retained relative to thc last lcarning trial. are presc'nted. Procedw'ul nremot\ On thc' Pursuit Rotor Task (Photoelectric Pursuit Rotor. Modcl 300I4; Lafiryette Instrunrent Cornpany. Lai'iryctte, Indiana). participants rverc rc'quircd to hold a curved light sensor wand with their right hand and track the motion of a îwo centimeter liglrt ernitting disk (the target) rotating insidc a 30.5 centirnctcr in diarnctcr tnmtablc. Participants initially cornpleted a baseline ofthrec trvcnty sccond trials at each of four ditlcrent disk rotation speeds ( I 5, 30, 45. and 60 rpm) and thc tilne-on-target rvas rccordccl. Thc avcrage tilne-on-târget for each ofthc thrce baselinc rotation speed trials was then calculated. For each participant. the rotation Derno_eraphic and clinical characteristics of the patient and control groups are presented in Table 1. The groups did not diflbr significantly in terms of age, ycars of education. scx composilion and handedness. The mean total score on the' l. Table Denrographic and clinical characteristics of OCD and control groups OC'D ( ()ntrol (n:21) (n - 29) MSDLISDT'p 37.19 I 1.2 I 35.14 9.83 0.71 .448 Measurc Ase Yearsofeducation 14.87 .3.91 16.38.1.66 1.,19 .l4l BDI 15.09 7.48 4.18 4.25 7.00 .001* STAI S 13.26 10.77 3?..14 9.71 4.06 .001* YBOC'S 'lbtal 28.33 YBOCS-Obsessions 13.71 YROC'S C'ompulsions Fentale (o,i,) 40.7 Right-handcd ('2,; il)egrees ol' fice dorn *p < 14.59 4.29 2.65 2.06 100 = 5,1 58.6 93.1 X2 0.18 0.29 0.49 0.26 tbr all comparisons, .05. llDl = Beck Depression ln\entory: S1Âl-S = Slale-l'raii An\iety lnventtrry-State subscale: \'BO( S : Yale-Bros'n Obsessive Conrpulsive Àot<,. Scal.'. 650 R.M. Roth et al. Tablc 2. Descriptive statistics lor pertbrnrance on procetlural OCD (n: 271 M Measure RAVLT (No. items recalled) Trial Trial Ttial t7.10 14.45 15.22 l 6.41 18.64 I 8.63 44.82 18.33 2.49 47.73 t7 .99 3.04 61.39 19.42 2.08 62.17 20.5.1 | .7 I 64.67 19.52 1.25 66.49 19.82 0.74 Trial t3.67 Pursuit Rotor Block I lllock 2 ( rl)egrees ol fieedonr */'< I l.4l 82.4t I 1.44 82.78 tinrc-on-targct) Block 3 []lock 4 Block 5 !ilock 6 tl 56.68 61.87 10.87 70.ti5 71.36 70.44 12.19 13.33 (Zr SD 14.00 3 Trial 4 5 lV .sD 7.07 2.05 0.22 10.28 2.20 0.28 12.07 t.77 0.26 13.00 I .81 0.73 3.03 2. 18 t.23 I 1.66 2.88 0.32 88.44 t2.42 1.46 I | .72 2.99 0.3(r 1i9.39 15.29 1.50 t4.21, l.l l 0.67 ll Short delay Short delay, perccnt rccall Long delay Long dclay. percent recall Recognition List A Conlrol = 29) 1n 1.45 2.t7 t.57 l .59 I .59 2.86 18.l.l 2.87 t] .67 1.27 6.96 t0. I 2 and dcclarative memory lests I .825 .779 .797 .469 .223 .748 .t49 .723 .140 .505 .016+ .004* .041+ .093 .215 .464 = 54 t'or all conrparisons. '05' YBOCS for the patient group was in the moderate symptom scverity rauge. Despitc excluding patients rvith a diagnosable dcpressivc disordcr, thc paticnt group rcported significantly more symptorns oldepression and anxiety than the control group. Descriptive statistics for the RAVLT and Pursuit Rotor Task are presented in Table 2. The mean number of rvords recalled on each learning trial of RAVLT is preser.rted in Fi-eure 1 for illustration. Neither the Group main elÏèct [f (1.,§4):0. 18. p: .671] nor the Group X Trial interaction IF(4,54) : 0.83.p : .5l0] reached statistical significance. As expccted, there was a significant main efï-cct of trial. indicating better recall with repeated exposure to the word list lF(4,54) :231.55. p: .0011. Additional analyses showed that the patients and controls performed similarly in the number ofrvords recalled and percent ofwords recalled at short and long delays, as well as on the recognition mem- ory trial. On thc Pursuit Rotor Task. the groups did not differ rvith respect to the rotation speed used for test trials, as determined by the baseline evaluation (X2 :0.203, p: .299). The majority of participants in both the patient (88.9%) and control (75.9%) groups perfonned the tcst trials at thc 60 ryrn setting, while the remainder used the 45 rym setting. Higure 2 shorvs the mcan percerlt time on-talget fbr the six test blocks of thc Pursuit Rotor Task. A significant main effect ID 14 12 10 of Group was present. indicating llrat. overall, the OCD group stayed on-target significantly longer than the control : 4.30. p : .0431. A main ef]ect of block group [F( I .54) was also observcd in<iicating bettcr pcrfbrmance with 12345 RAVL'l"hial *ü** OCD "'^ '' Control Fig. l. N{ean Test by group. wordspertrial on the ReyAuditory Verbal Learning increasing task experience across the groups If'(5,54) : 46.92. p: .0011. The Group x Block interaction was also significant [.F (5, 54) : 2.77 , p : .019]. Group comparisons on thc individual test block performance indicated that the OCD group stayed on-target significantly longer than controls for the first three blocks. Groups did not differ on the last three blocks. Statistical comparison of patients with (n : 8) and without (r = l9) a comorbid anxiety disorder did not reveal any Menu»v in OCD 651 must be cxertcd in linking neuropsychological pertbr- 80 mance changes 1o specilic underlying neural circr.ritry, tentative support {br this hypothesis may be found in research on brain regions involved in early lersu.r later stages of procedural lnemory tasks. Studies rvith primates and healthy humans indicate that the caudate nucleus aud rostral putamen tend to be most active during the carly stages of novel ploccdural mernory tasks ( Krebs et al., I 998; Miyachi et al., 2002). In contrast. later stages oflearning appear to involve corticocerebellar circuitry and rnore caudal portions ofthe putalnen (GralIon et al.. 1994: Krebs et al.. 19981 Miyachi et al., 2002). Horvcver, it shoul«l be notcd that neither regional overactivation nor underactivation on furrctional neuroirnaging are consistently associatcd rvith perfbrmance differenccs ou a variety of tasks in cither hcalthy controls or 75 70 65 60 55 50 45 40 12345b Trial tslock -'-l--OCD Fig. 2. ''a-" Control Percent tirnc on larget on the Pursuil Rotor Task by groulr patient populations (e.9.. McAllister et al.. 2001: Rauch al.. 1997.1. Only trvo functional neuroimaging studics to date have applied proccdural mclnory tasks to OCD, both with small et sarnple sizes. One snrdy used PE,T and an irnplicit sequence significant difÈrenccs or1 either the RAVLT or Pursuit Rotor Task (allp > .05). No significant comelation (allp > .05) u,as obselvcd betwecn percent tirne ou target for any ofthe Pursuit Rotor blocks an«.l overall OCD syrnptom scvcrity (YBOCS total score), or sevcrity of obsessions (YBOCS Obsession subscale) or compulsions (YBOCS Cornpulsion subscalc). F'inally, group diflèrences in mood did not accottt.tt fbr thc Pursuit Rotol Task findings in analysis of covariancc-. DISCUSSION Thc prescnt study rcvc-alcd that uurnedicatcd paticnts with OCD pertbrmed better than hcalthy controls on a procedural memory task. while group differences were ltot observed tbr either encoding ol retricval in declarative rnernory. Specitically. thc OC'D paticnts l'cnraincd on-target during the' Pursuit Rotor Task for a sigrrificantly gl'cater percent of tirne overall thau controls. This effect rvas largely dr-rc to significantly bettcr pcltbrrnancc by the patients during the carly but not latcr trial blocks of the- task. No association was fbund bet*'ecn procedural rnemory and rnood or the scverity of eithcr obsessions ol compulsions in thc patient group. 'fhat our OCD group shorved bettcr procedural tnetrory than controls, especially during the early block of trials, was uncxpccted. Patients rvith Tourctte's syndrome, a disordcr hypothcsized to bc genctically rclated to OCD (Sheppard et al.. 1999;, have been shorvn to pcrtbrrn rnore poorly than healthy controls orr the Pursuit Rotor Task (Stebbins et al.. 1995). Howcver. iu coutrast to OCD. Tourette's syndronrc is associatcd rvith striatal underactivation (llraun et al.. 1993; E,idclberg et al.. 1997). tt may be speculated that enhanccd procedural lnemory during the early stages of learning in OCD is associatcd rvith thc striatal ovelactivation reported in nunrerous firnctional neuroimaging studies of the disorder (Saxer.ra & Rauch, 2000). Although caution learning task (Rauch et al., 1997). r,r'hile the other (Roth et al.. 2003) employed f'unctional MRI and a probabilistic classification task (Knorvltou ct al., 1994;. Despite the use of different tasks, both studies reported that adults with OCD showed prorninent hippocampal but little striatal activation, while healthy controls showed prorniner,t striatal and little to no hippocampal activation. Perfonnance differeuces were not observed in either study. These results were interpreted as consistent rvith fiontostriatal dysfunction in OCD. Sincc neithcr of thc studics spccifically reported on temporal changes ir-r the degree of striatal activation during the course ofthe tasks, the hypothesis that diff'erential striatal activalion in OCD is related to the carly stage of procedural mcnlory tasks could not be addressed. Taken together. results frorn neuroirnaging studies dur'ing rest, sylnptom provocation and procedural metnory tasks suggest that the directionality of thc striatal abnormality seen in OCD. whether it be overactivation or underactivation, may depend on the stage of procedural memory interrogated and/or the naturc of the task ernploycd. With respect to task requirerncnts. both sequence learning and probabilistic classification involve "cognitive" procedural mcmory. ln contrast. the Pursuit Rotor Task involves a niore pulely "motor'" procedural n'lcmory. Activation of thc ltippocalnpus durirrg cognitive procedural memory tasks in OCD has been interpreted as suggesting that these patients have abnormal tiontostriatal circuitry and theretbre rnust rcly on thc declarative lncmory systcm to adequately perform procedural rnernory tasks (Rauch & Savage,2000). lt rctnaius unknown rvhether patients with OCD would also activate the hippocampus. or show diilèrential striatal activation. during a motor proccdural mcmory task such as that employed in the present investigation. Further functional lucuroimaging studies of OCD, ernploying both cognitive and rrotor procedural mcmory tasks, and assessiug activation both during the early and later stages ofthe tasks. will be necessary to evaluate these hypothcses. R.M. Roth et ul. 652 Thc f'unctional signilicancc of better procedural mcmory in our paticnts is nnclear. One aspect of procedural lnemoty subserved by lrontostriatal circuitry that lray be relevant to OCD is thc fbrrnatiou ol stirnulus-rcsponse associatious ( Mishkin ct al.. 1984; Packard & Knorvlton, 2002; Poldrack ,l995). ln the casc of OCD. this et al., 2001; Saint-Cyr et al., rnay be exprcssed as thcilitated acquisition ofassociations bctwccn potcntially threatcnine cogrtil ive- ol cnvirorttnental stirnuli such as irrtrusive thoughts ar-rd anxiety-reducing conrpulsive behaviors. Exposure and respouse prevention, a fotm of behavior thcrapy that is hi-ehly ctï'ective fbr trcating OCD. involves irr part repcated presentation of anxicty-provoking stirnuli and gradual rvittrhotding of compulsive responscs (Foa et al. 2001). This treatrncnt. rvhich has been shown to rcduce resting t'r'or-ltostriatal overactivation in patients rvith OCD on positron emission tornography scans (Schwal'tz. 1998), may rvork in part through the breaking of maladaptive stimulus-rcsponsc associat ions. Our OCD and control groups did not cliller rvith rcspect to encoding and rctrieval in declarative tnernory. Prior research has yielded inconsistent findings lbr both verbal and r.'isual declarative mclnory in the disorder (Grcisberg & McKay. 2003). Rcccnt studics havc indicatcd that lesser use oforganizational strategies during encoding. rather than inTpailed retricval processes, accourlts lbr ve-r'bal and visual declarativc memory dellcits rvhcn observed in OCID (Deckersbach ct al.. 20001 Savage ct al.. 1999). This has been interpreted as ret'lecting a deficit in the use of frontostriatal circuitry rnediatcd stratcgic processes rather than a ntesial tcrnporal circuitry rnediated retrieval deficit lSavagc et al., 1999). The lack of gloup dit-fèrence for declarative memory in the plesent study may thus be due in part to the natule of thc RAVLT. Unlike sornc othcr declarativc mcrnoly tcsis such as the ('alifbntia Verbal Leaming Test (CVLT; Delis et al., 1987). the iterns on the RAVLT cannot be readily organized on thc basis of semantic or other relationships beîrn'cen iterns, and thclelble thc potential advantage in recall afforded to controls rvas likely limited. L,vcn if a test such as the CVLI had been ernployed, previous findings (Savage et aI..2000; suggcst that group diflerences in learning and rccall rvould likcly lravc been due to thc paticnt groups lesser usc of strategic proccsses rather than a retlicval def- icit per se. The prescnt tindiugs colltrast with tlrc only prior study using the Pursuit Rotor'l-ask in OCD, u,hich found no differences betrveen healthy controls and patierlts ( Martin et al., I 993 ). ln attelnpting to account for their results. Martin and collcagucs proposcil that thc high [cvcl oleducation ofthcir participants (Â1 age OCD : 15.4: içl age controls : 15.9 years) may have rendered it difficult to obse'rve group diffèrences. Horvever, thc level of education of participants in thc prcscrlt study is cornparable to that leport. Furthennole, both studies uscd unmedicated patients, and treatment rvilh selective serotonin rcuptake inhibitors does not appear to havc a significant irnpact on procedural lucmory in OCD (Mataix-Cols ct al.. 2002). Thus. educati<m and tredication status are unlikely to explain the discrepancy in findings. lt is possible that thc larger samplc size employed in the cur- rent stucly may have contributcd to the detection of group differences. It may be questioned rvhether the better procedural memory by our patient group is an artifàct ofpoor perfbnnance by the control group. Howevcr, the percent time-on-target for our controls during all stages ofthe Pursuit Rotor Task is cornparable or better than that observed in scveral other studies using healthy controls rvith similar agcs arrd educa- tiorr levels (Martin et al., 19931 Stebbins et al., l995: van Gorp et al., 1999). It may also be qr.restioned whether the comparison of tests that employ diilbrent stimulus modalities. auditory.-verbal ( RAVLT) r'ersls visuomotor ( Pursuit Rotor), r.nay have impacted our findings. The methodology and results ofthe present study do not perrnit resolution of this issuc. Further studies employing both auditory-verbal dcclarative lrernory tests arrd auditory-verbal procedural memory tests such as prir-ning tasks rvould help address this issue. Our OCD group shorvcd grcater time-orl-target than con- trols as early as the flrst block of test trials on thc Pursuit Rotor Task. This raises the possibility that the patients had greater skill than controls at baseline. Our groups did not differ significantly in the rprn levcl at rvhich their ability to stay on target was closest to 5 s during baseline trials: this has been thc standard n-rethod to cstablish baseline equivalcncc in rnotor abilities on the task (Heindel ct al., 1989; Stebbins ct al., 1995; van Gorp et al.. 1999). Horvever. significant group differeuces in the rate of improvernent in actual time on target during the baseline trials rnay have resulted in the patient group starting the tcst blocks rvith an already higher level of pertbrmance. We therefore analyzed the time on target for the last baseline trial to gauge level of perfonnance prior to the first tcst block. This was fourrd to be significantly bettcr for the patient group [t(l):2.18, .034). lt is possible that the nrultiple baseline trials were sutïcient to pennit group dilterences in procedllral memory to manifcst, which rvere thcn maintaincd until the fourth block rvhen the patient group reached a plateau in their rate of inrprovement. F uture studies would benefit from using sizably larger sarnples in ordcr to permit statistical cvaluation of thc ratc of lcarning during both thc baseline p: and test phases of the task. Altemately. beginning test blocks at the same rpm rate tbr all subjects and elirninating pre-test block basclinc trials altogether may bc hclpful fbr detcrmining differenccs in procedural rnemory in early and late phases rvithout extensive pre-exposure to the task (e.g., Cranholrn et al., 1993). In surrmary, the prescnt findings revealcd signiticantly better procedural memory duling the early stages of lcarning in unrnedicated patients rvith OCD as compared to healthy controls. in the absencc of group differences for cncoding and rctricval in declarativc mclrrory. This pattern of findings appears to be consistent rvith evidence of frontostriatal circuitry abnormality in OCD. Replication in a largcr sarnplc of paticnts with OCD. and direct comparison with patients rvith Tourette's syndrome, rvould be infonnative. Finally, further lirnctional neuroimaging studies would be hclpful to directly cvaluate the relationship between cog- 653 Lletnor.r,in OCl) nitive and rnotor procedural mcrnory task pcrlbrmancc and frontostriatal circuitry activation in OCD. ACKNOWLEDGMENTS Psychological Corporation. This study \vas supported by grants tiorr the F'onds Pour La Forrnation de Chercheurs et L'Aide a la Recherche (l{MR). and the Fond cle la Recherche en Santé Québec (KO and JB). We thank Laura F'lashnran and three anonyrnous revic-wers for their helplul comrnents on this manuscril'rt. REFERENCES Anrir. N.. l-reshrnan. ÿ1., & Foa. I"-.8. (2000). Farnily distress and involvernent in relatives of obscssive-contpulsir,'c disorder patients. Journol ol .4ntiet.v Disorders. 14,209-217. Arnerican Psychiatric Association. (19911. Di ugno.slic' nd sloti.sti<'ul munuol ol mentul disonlcrs (4th ed.). Washington, DC: Arne-rican Psychiatric Association Press. u Aylrr,ard. E.ll.. Ilanis. & G.J., lloehn-Salic. R., Barta. P.E.. N{achlin. Peallson. G.D. (1996). Nortlal caudate nucleus in obsessivc-conrpulsir'e <lisordel assessed by quantitative neuroirnaging. .4rchives ol Getrcral Ps.t'chiutrl'. -r3.577 584. Bcbbington. P.lr., ( I998). Epidemiology of obsessive-cotnpulsive S.R.. disorder: Brirish,lourna\ d Ps1'c'hiur.t' Supp|ementun (35),2 6. Beck. A.T. (1987). BecÀ Depression lnventory. San Antonio. TX: The Psychologieal ('ttlporation. Braun. A.R.. Stoetter. 8.. Randolph. C.. Ilsiao. J.K.. Vladar. K.. Gernert..l.. ( arson. R.Fl.. Ilerscovitch. P. & Chase. T.N. ( 1993). Thc f'unctional Neuroanatorny ol Tourcttc's Syndronrc: An F DG-PET Study. I: Regional change's in cercbral glucose ntetab- olisnr dilierentiating patients and controls. Neurops.r'chttphur- mutolog.t,.9. l5l 168. Brown. T.A.. Di Nardo. PA.. & Barlo*. D.H. (1994). Antiery disonlers inten'iev'schedule.li» DSI'{ ly. Albany, NY: Graywind Publications. Bylsnra. F'.W.. Branclt. J.. & Strauss. M.E. (1990). Aspccts of plocedural urenror! arc diTterenlially inrpatrcd in Iluntington's cliscase. .4rz'lilr,r,s o/ C' Ii ni ca I N artntps.t'choIo91, 5. 287 297. Bystritsky. A.. [-iberman. R.P, Hwang. S.. Wallacc. C.J., Vapnik. T., Maindnrent. K.. & Saxena. S. (2001). Social tunctioning and quality ollilè conrparisons bctwecn obscssive-cotnpulsivc and schizophrenic disorders. Dt'pressiott ttnd An.tiety. 14, ll4 218. C'abeza. R. & Nybcrg. L. (2000). Inraging cognition Il: An eurpirical rer'icw o1275 PF,T and {'lvlRI studics. Jorrrrrrl of Cognitnt' Neuroscientc. l:. | 17. Cavcdini. P., (lisinra. lvt.. Riboldi, (i.. D'Annucci. A.. & Ilcllodi. L. (200 I ). A neuropsychological sturly of dissociation in cor- tical antl sr.rbcortical firnctioning in obsessive-cotnpulsive dis- order by Toiler 357 363. of IIanoi with obsessivc-conrpulsive disorder. .4nrerican Jourttul oJ Psyc'hiatr.y, 159. 1780 1182. Delis, D.C.. Kramer. J.H.. Kaplan. E., & Ober, Il.A. (1987). &r/i/brniu l/erbul Leurning Test Adult lâr.srort. New York: The tâsk. Bruin anrt C'ognition, 46, Christenscn. K.J., Kirn, S.W.. Dyskcn. IV.W., & Hoover, K.M. obscssirc- in Psychiatr.v. J/.4 (1992). Ncuropsychological perfortnance 18. cornpulsivc disorder. Biologicol Deckersbach. T.. Olto. M.W., Savage. C.R.. Baer, L.. & Jenike, M.A- (2000). The relationship lretween senrantic organization and rnernory in obsessive-cornpulsi ve disorder. P,s1't' lntheru pl' unl Pst<ltoso»rutit's, 69. l0l l()7. Deckersbach. 'f.. Savage. C.R.. Curran. T.. Bohne. A.. Wilhelnt. S.. Baer. L.,.lcnike. N'I,A.. & Rauch. S.L. (2002). A study of parallel implicit and cxplicit inf-orntation processing in putients Eidelberg, D.. Moeller. J.R,. Antonini,4., Kazumata, K.. Dhawan. V.. Budnran. C.. & Feigin. A. (1997). The rnetabolic anatomy ol Tourelte's syndronre. N e u ro lo g1'. 4 B, 927 934. F'oa. E., Wilson, R., & Wilson. R.R. (200 I ). Stop obsessing!: How îo overconte _rour ohse.ssiotts and compul.:iotts ( Rev. ed.). New York: Bantanr Doubleday Dell Publishers. Gabrieli. J.D. ( 1998). Cognitivc neuroscience of hurnan melrory. .4tnuul Review of Psycholog.r', 49,81 ll5. Gabtieli, J.D., Stebbins, G.T.. Singh, J.. Willinghanr. D.. & Goetz, C.G. ( 1997). lntact mirror-tracing and impaircd rotary-pursuit learning in patients with lluntington's disease: Evidence for dissociable rnel.nory systems in skill leaming. Neuropsl;<'hol- ogv. 11.272'281. Gonrez Beldarrain, M., Grafman. J.. Pascual-Leonc. A.. & GarciaMonco. J.C. ( 19991. Procedural learning is irnpaired in patients with prciiontal lesions. À'erln1rlg1,.52. I853 I8(r0. Goodnran" W.K.. Price. L.H., Rasnrussen. S.A., lv'lazure, C.. Delgado. P.. Ilcninger. G.R.. & Charney. D.S. (1989). The YaleBrorvn Obsessive-Cornpulsil'e Scalc. I: Developrncnt. use. and reliability. Archives of Gene rol Psy'chiatr.t',46. l0l2-1016. Grafton, S.T., Woods. R.P. & Tyska, M. (1994). Functional imaging ofprocedural niotor leaming: Relating cerebral blood flow with individual subject perlbnnance. Ilumun Brain lllopping. t.221 214. Granhohr. Fl.. Éiartzokis. G.. Asamou R.F'.. & Marder. S.R. ( 1993 ). Prclirninary associations betlvcen rnotor procedural learning. basal ganglia T2 relaxation times. and tardive dyskincsia in schizophrenia. Ps)'cltiatry Reseorch. 50. 33 44. Greisberg. S. & McKay, D. (2003). Neuropsychology ofobsessivcconrpulsive disorder: A revierv and trea{lnent irnplications. C/il- icol Psycholog.tt Review. 2 3, 95 I 17. Heindel" W.C.. Butters, N.. & Salnron, D.P (1988). lmpaired leaming of a nrdor skill in patients rvith Huntington's disease. r9cftar. ioral Neuroscien<:e, 102. l4I 147. Fleindel, W.C--.. Saluron. D.P.. Shults. C.W.. Walicke. PA., & Butters. N. (l9ti9). Neuropsychological evidence for nrultiple inrplicit nlelnory systerns: A cor:rparison of Alzheinter's, lluntington's, and Parkinson's disease patients. Journal ol Ne urosc ien<'e. 9. 582-587. Hollander, E.. Schillman, E.. Cohen,8.. Rivcra-Stein, M.4.. Rosen. W., Gorrlan. .1.M.. Fyer, A..1.. Papp. 1... & Lietroivitz. M.R. (1990). Signs of ccntral nervous systenr dyslunction in obsessivc-compulsive disordcr.,!rc'ltives of Generu| Ps.t'chiutry, 17.27 32. Horwath, I-.. & Weissrman. lvl. M. (2000). Thc epiderniology and cross-national presentation of obsessive-compulsive disorder. Ps.t,chiatric Clinics tl North ,4nterica. 23.493 507. Kalechstein. A.D., IIinkin. C.H.. van Gorp, W.G.. Castcllon, S.A.. & Satz. P ( 1998). Depression predicts procedural but not episodic nrerrory in HIV- I infection. Jotrrnutl o/'Clinitul uud Experinentu I Neun4tsycholog.r'. ) 0. 529 535. Knowlton. 8.J.. Squire. L.R.. & Gluck, I\'1.A. ( 1994). Probabilistic classitlcation learning in arnnesia. Leorning ond Menrory,, l. t06 120. Krcbs. ll.l.. lirashcrs-Krug.'f.. Rauch. S.L.. Savagc. C.R.. Hogan. N.. Rubin, R,ll.. f-ischrnan. A.J.. & Alpert. N.M. ( 1998). Robotaided firnctional irnaging: Application to a motor learning study. Human Broin lllapping,6,59 72. R.M. Rorh et al. 654 Martin, A.. Pigolt. T.A.. L:rlonde. F-.M.. Dalton. I.. Dubbert. 8.. & Murphy. D. L. ( 1993 ). Lack of evitlence lor Huntington's diseaselikc cognitive dysfunction in obsessivc-conrpulsive disorder. lliologit:ul Psychiurrv, -?J, 345 353. lvlataix-Cols. D., Alonso, P. Pifarre, J.. Menchon, J.M.. & Vallejo, l. (2002l'. Neuropsychological perlbrnrance in rnedicated vs. unrnedicated patients with obsessive-colnpulsive disorder. Ps_rh io tt'.t' Re s eo rt' h, I 09. 255' 264. McAIlistsr. T.W., Sparling, M.8.. I-lirshman. L.A., Guerin. S..1., Ivlarnourian. A.C.. & Saykin. A.J. (2001 ). Differential r.r'orking nrenroly load effccts al'ter milcl trauuratic brain injury. Neuroinrage. 14. I004 1012. Mishkin. M.. Ivtalamut. 8.. & Bachcvalicr, J. (1984). Iv{emolies <' and habits: Trvo neural systerrts. In G. Lynch, J.L. McGaugh. & N.lvl. Wcinbcrgor ( Eds.). Neurobiolog.t, ol learning; atrul Dten- o/r'(pp. 287 296). Ncw York: Guilf'old Press. Mi1,achi, S., Ilikosaka. O.. & Lu. X, (2002). Diffèrential activation of nronkey striatal neurons in the early and late stages ol' prcrcedural learning. Experincntttl Bruin Reseurch. 146. 122 t26. Packard. I\4.G. & Knowlton, ll..l. (2002). Lcarning and nrcrnory frrnctions ofthe basal ganglia. ,lnnunl Revie:l ol Neuroscit'nce, l-r.563-593. Pascual-Leone, A.. Gralirran. J., & IIallett, IVI. (1995). Procedural Iearning and lrretiontal co(lèx. .4nnals fi the Neu' htrk Acadenty of Sciences. 769.61 70. Poldrack. R.A.. Clark, .1." Pare-tllagocv. ti.J.. Shohamy, D., Cleso Moyano. J., Myers, C'.. & Ciluck. N'I.4. (2001). lntcractive rnemory syslcrl.rs in the huuran brain. Norure. 411. 516-550. Rauch. S.L. & Savage. C.R. (2000). lnvestigating cottico-striatal pathophysiology in obsessive-cornpulsive tlisorders: Proce- dural lcarning and irnaging probes. ln Vy'.K. Coodrnan, M.\'. Rudorfèr & J.D. Maser ( [:ds. ). O|.rre.s.si le-compulsive tlisonler (pp 133 154). Mahwah, NJ: Lai.vrence Erlbaurn Associates. Inc. Rauch. S.L.. Savagc. C'.R., Alpert. N.M., Dougherty. D.. Kendrick. A.. Curran. T.. Brorvn. ll.f).. Manzo" P. F-ischrnan. A.J.. &.lcnike. M.A. (1997). Probing striatal lunction in obscssiveconrpulsive disorder: A PliT study of'iniplicit sequcnce learning. Jourrrul of Neurops.vchiatr.t' arul Clinicul Neuroscicnces, 9.5(18-571. Rauch. S.l-.. Whalcn, P.J.. Curran. T.. Shin. L.ivl.. (loffey. B.J.. Savage. Ci.R.,lvtclnerney, S.C., Baer, L.. &.lenike. M.A. (2001). Probing striato-thalamic f unction in obsessir,'e-conrpulsive disordcr and Tourette syndronre using neuroinraging nrcthods. -4dt'Ltnce,s in Neurolog.r,. I 5. 207 224. Robirrson, D., Wu. I I.. lVlunne, R.A.. Ashtari. ivl.. Alvir, J.M., Lemer G.. Koreen. A. Colc. K.. & Bogcrts. B. (199-5). Reduced caudate nucleus volume in obsessive-conrpulsivc disorder. lrcltives ol Generttl Ps.r'<hiatrl', Jl. 393 398. Rosenberg. D.R.. Keshavan. M.S.. O'llearn. K.M.. Dick. E.L.. Bagwell. W.W.. Seynrour'. A.8.. It{ontrosc, D.M.. Picrri. J.N., Sainl-Cyr, .1.A.. Taylor. A.E.. & Nicholson, K. (1995). Behavior and thc basal ganglia. Atlvances in Neurolog.t,,65, I 28. Sarazin, M.. Deweer. 8.. Merkl. A.. Von Poser. N.. Pillon. 8., & Dubois. B. (2002). Procedural learning and striatofrontal dysfunction in Parkinson's disease. ,ÿlovement Disorders, 17, 265 273. Savage, C.R.. Baer, L., Keuthen, N.J.. Brorvn. H.D.. Rauch. S.L., & Jcnike. M.A. (1999). Organizational strategics nrediate nonverbal r.nernory irnpairment in obscssive-compulsive disorder. Biological Ps1'chiatry.4j, 905 916. Savage. C.R.. Deckersbach. T., Wilhelnr, S., Ratrch. S.L.. Baer, L., Reid. T., & .lenike, M.A. (2000). Strategic processing and episodic nrenrory inrpairnrent in obsessive compulsive disordcr. Neuropsycholog.r,. 14. l4l l5l. Savage. C'.R.. Keuthen. N.J.. Jenike. M.A., Brown, ILD.. Baer. L.. Kendrick. A.D.. Miguel. E.C.. Rauch. S.L., & Alpert. lv4.S. 996). Recall and recognition nlelrory in obsessive-compulsive disorder. .kturnul of Neurops.w:hiatry uncl CIinic'ttl Neuros<:i( 1 ent'es.8.99 103. Saxena, S.. Bota, R.G.. & Brody, A.L. (2()0 11. Brain-behavior relationships in obsessive-compulsive disorder. Senrinors in Clinical Ncutopsl'chiatr.r,. 6. 82 l0l. Saxena. S.. Brotly.4.L.. Ho, M.L., Alborzian. S.. Maidnrent. K.M., Zohrabi, N.. Ho. M.K. FIuang. S.C., Wu. H.M., & Baxter, L.R. (2002). Differc-ntial cerebral rnetabolic changes with paroxetine treatlnent ofobsessivc-compulsive disorder vs. major depression. ,4irliives o/ General Ps.t'chiatr.r,^ 59,250 261. Saxentr. S. & Rauch. S. L. (2000). Functional neuroinraging and the neuroanatolry ol'obsessive-compulsivc disorder. Psychiatric' Olinics ol Norrh .lntericu.2J. 563 586. Schwartz, .1.M. (1998). Ncuroanatornical aspects of cognitivcbehavioural therapy response in obsessive-corripulsive disor- der. An evolving perspective on brain and behaviour. Bnllslr Journal oJ Psla'hiqlvy 5upplemenîum (35), 38-41. Sheppard. f).M.. Bradsharv. J.L.. Purcell. R.. & Pantelis. C. (1999). Tourctte's and conrorbid syndromes: Obsessive conrpulsive and attention deficit hyperactivity disorder. A conrmon etiology'l Clinical Psvcholog.r' Àelleu', 19.531 552. Skoog. G. & Skoog. I. (1999). A 40-year follow-up of patients n,ith obsessive-compulsive disorder. .4rchiyes ol GeneruI Ps1- chiittry, 56, l2l-127. Spielbergcr, C.. Montouri. F-.. & Luschene. R. (1970). Stote-Troit An.riery lnvenror.l, (STAI l"orm C I ). Palo Alto. CA: Consulting Psychologists Press. & Strauss" E. (1998). A conpeniliunt ol neuropst,chologicol tests (2ntl ed.). New York: Oxfbrd University Press. Squire. t-.R. (1986). lVlechanisms ol menrory. Sr:ience. 2.]2. Spreen, O. I 612 l6l 9. Squire. L.R. &Zola, S.M. (1996). Structureand functionofdeclarative and nondeclarative memory systems. Ptoceedings of tlte National .4caclenrv ol St'ie nces ol the United States of Anrcrita. 9J. t3515 13522. & Birmaher, B ( I 997 ). Frontostriatal lueasurernent in treâtmentnaive childrcn rvith obscssive-conrpulsive disorder. .4rchives ol Generul Ps.t'chiatr.t.54. 824 ti30. Roih, R.M.. Saykin. A.J.. Wishart. l'1.A." F-lashrnan. L.A., Ward. J.. & Marnourian. A.C. (2003 ). An flvlRI study ot'brain actir ation during irnlrlicit learning in OCD. Journul oJ Nerntps.lchiutry Stebbins, G.T.. Singh. .1.. Weiner. J., Wilson, R.S.. Goetz. C.G., & Gabricli. J.D.E. ( 1995). Selective impairmcnts of rncmory functioning in unnredicated adults rvith Gilles de la Tourette's syndrorne. Àtro o7r.ry'c hologl', 9, 329 337 . Szeszko, PR., Robinson. D.. Alvir. J.lr'1.. Bilder, R.M.. Lensz. T.. Ashtari. M.. Wu. H.. & Bogerts. B. ( 1999). Orbital lrontal and atul Clini<'ul Neuro,tc ien<'es, I 5, 262. Sabc, L.. Jason. L., Juejati. NL. Lciguartia. R., & Starkstein, S.E. ( I 995). Dissociation betrvecn dcclarative and procedural leârning in derrrcntia and dcpression. Journttl ol Clinical und Expcr- anrygdala volurne reductions in obsessive-compulsive disordcr. .4rcftilc,s ol Generul Ps.tchiatn', -16. 913 919. van Gorp. W.C.. Altshuler, L.. Theberge, D.C.. & Mintz, J. ( 1999). Declarative and procedural melnory in bipolar disorder. Biologic a I Ps.t <:lt iatry. 4 6, 525 -53 l. inerial Nettropsrchologv. 1 Z. 841 84t1.