European Neuropsychopharmacology (2013) 23, 790–798

0924-977X/$ - see frhttp://dx.doi.org/1

nCorrespondence tE-mail address: p

REVIEW

Treating impaired cognition in schizophrenia:The case for combining cognitive-enhancingdrugs with cognitive remediation

Panayiota G. Michalopouloua,n, Shôn W. Lewisb, Til Wykesc,Judith Jaegerd,e, Shitij Kapura

aSection on Schizophrenia, Imaging and Therapeutics, Department of Psychosis Studies,Institute of Psychiatry, King's College London, UKbInstitute of Brain, Behaviour and Mental Health, University of Manchester, Manchester, UKcDepartment of Psychology, Institute of Psychiatry, King's College London, UKdCogState, New Haven, CT, USAeDepartment of Psychiatry and Behavioral Sciences, Albert Einstein College of Medicine, NY, USA

Received 14 December 2012; received in revised form 26 February 2013; accepted 8 March 2013

KEYWORDSSchizophrenia;Cognitive impair-ment;Cognitive-enhancingdrugs;Cognitive remedia-tion;Combination of cogni-tive remediation withcognitive-enhancingdrugs

ont matter & 20130.1016/j.euroneur

o: PO Box 053, Instanayiota.michalo

AbstractCognitive impairment is a well-documented feature of schizophrenia and represents a majorimpediment to the functional recovery of patients. The therapeutic strategies to improve cognitionin schizophrenia have either used medications (collectively referred to as ‘cognitive-enhancing drugs’in this article) or non-pharmacological training approaches (‘cognitive remediation’). Cognitive-enhancing drugs have not as yet been successful and cognitive remediation has shown modestsuccess. Therefore, we may need to explore new therapeutic paradigms to improve cognition inschizophrenia. The optimal approach may require a combination of cognitive-enhancing drugs withcognitive remediation. We review the available data from animal and human studies that provide theconceptual basis, proof-of-concept and illustrations of success of such combination strategies inexperimental and clinical paradigms in other conditions. We address the major design issues relevantto the choice of the cognitive-enhancing drugs and cognitive remediation, as well as the timing andthe duration of the intervention as will be relevant for schizophrenia. Finally, we address thepractical realities of the development and testing of such combined approaches in the real-worldclinical situation and conclude that while scientifically attractive, there are several practicaldifficulties to be overcome for this approach to be clinically feasible.& 2013 Elsevier B.V. and ECNP. All rights reserved.

Elsevier B.V. and ECNP. All rights reserved.o.2013.03.012

itute of Psychiatry, King’s College London, De Crespigny Park, London SE5 8AF, UK. Tel.: +44 2078480356.poulou@kcl.ac.uk (P.G. Michalopoulou).

791Treating impaired cognition in schizophrenia

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7912. Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 791

2.1. Cognitive remediation approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7912.2. Cognitive-enhancing drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7922.3. Rationale for combining cognitive-enhancing drugs and cognitive remediation . . . . . . . . . . . . . . . . . . 7922.4. Design issues and practical realities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7942.5. Which cognitive-enhancing drugs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7942.6. Which type of cognitive remediation? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7952.7. When and how to combine cognitive remediation with cognitive-enhancing drugs? . . . . . . . . . . . . . . . 795

3. Practical issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7964. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796Role of funding source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 796

1. Introduction

In the earliest accounts of schizophrenia Kraepelin,who coined the term dementia praecox, and Bleuler,who revised it to schizophrenia, both emphasised theimportant place of cognitive impairment in defining theillness. Since then, hundreds of studies have documen-ted marked impairments across many cognitive domains(Dickinson et al., 2007). Cognitive impairment asso-ciated with schizophrenia (CIAS) is well established bythe time of the first episode; it shows limited relation topositive psychotic symptoms and is largely unaffected byantipsychotics (Reichenberg and Harvey, 2007). CIAS isfirmly linked to functional outcome with effect sizes (ES)of the association in the medium range for individualcognitive measures (e.g. Cohen's d from 0.41 to 0.87)(Green et al., 2000) and a larger magnitude associationwhen composite cognitive scores are used (e.g. Cohen'sd from 1.42 to 2.41) (Harvey et al., 1998). This associa-tion acquires particular significance when one recog-nises that despite good remission of psychosis,functional recovery in terms of gainful employment,independent living and managing social relationships isquite poor.

Most efforts to improve cognition in schizophrenia haveeither attempted pharmacological compounds (collectivelyreferred to as ‘cognitive-enhancing drugs’ in this article)or non-pharmacological training approaches (collectivelyreferred to as ‘cognitive remediation’ in this article). Bothapproaches aim to improve cognition and ultimately improvethe functional outcome of schizophrenia.

We first comment on the current evidence of the stand-alone efforts with cognitive remediation and cognitive-enhancing drugs and question whether they are sufficientlyoptimal. We then provide the scientific background, theproof-of-concept trials that have combined the two mod-alities and the practical and regulatory challenges in doingso in schizophrenia.

2. Background

2.1. Cognitive remediation approaches

A variety of training interventions to improve cognitiveabilities in patients with schizophrenia have been devel-oped ranging from teaching patients how to improve theirperformance on a single neuropsychological test to provi-sion of a comprehensive broad-based clinical remedia-tion program. The most recent meta-analysis of the con-trolled studies of cognitive remediation in schizophreniashowed a modest improvement in overall cognitive perfor-mance (Cohen's d=0.45), with some durability of effects,as shown in follow-up studies (Cohen's d=0.43) (Wykeset al., 2011).

Regarding the neurobiological basis of these improve-ments, the majority of the relevant data come fromfunctional neuroimaging studies on working memory train-ing, where performance enhancement is associated withactivation changes in cortical neural circuits implicated incognition, i.e. increases in prefrontal cortical function inpatients with schizophrenia, leading to a pattern of neuralresponse closer to the pattern seen in untrained healthyindividuals (Wykes et al., 2002). Cognitive remediation hasbeen associated with neuroprotective effects against greymatter loss in temporal brain regions associated withcognition (e.g. hippocampus, parahippocampal gyrus)(Eack et al., 2010) and also with increase of serum brainderived neurotrophic factor (BDNF) levels (Vinogradovet al., 2009).

Despite this encouraging evidence, there is a number ofstudies with negative results (e.g. Dickinson et al., 2010)and even where results are positive not all translate intofunctional benefits. There are however stronger effects onfunctioning when cognitive remediation, and especially astrategic remediation approach, is provided with someother form of psychiatric rehabilitation (Wykes et al.,2011). Cognitive remediation is not as yet recommended

P.G. Michalopoulou et al.792

as a treatment for CIAS by two recent guidelines (Dixonet al., 2010; NICE, 2009).

2.2. Cognitive-enhancing drugs

Various classes of compounds have been proposed to berelevant to CIAS by the Measurement and TreatmentResearch to Improve Cognition in Schizophrenia (MATRICS)initiative. These include cholinergic (especially nicotinicalpha-7 agonists), dopaminergic, glutamatergic agents,alpha-2 adrenergic receptor agonists, agents acting on theγ-aminobutyric acid (GABA) system and on serotonin recep-tors (Buchanan et al., 2007). Other compounds suchas cannabidiol and cannabinoid receptor antagonists, H3receptor antagonists, modafinil and also compounds withneuroprotective properties such as neurosteroids (e.g.pregnenolone and dehydroepiandrosterone) (for reviewsee Barch, 2010) and erythropoietin (Ehrenreich et al.,2007) have also been tested for their effects on CIAS.

Until April 2011, 17 controlled trials of add-on potentialcognitive-enhancing compounds were identified (Keefeet al., 2011). The majority of these trials were under-powered to detect small and medium ES, increasing there-fore the risk of not detecting a compound's cognitiveeffects. While it is difficult to draw general conclusions onthe basis of the limited available evidence, some cognitive-enhancing drugs have shown promising results for CIAS (e.g.minocycline, modafinil, galantamine), where they generallytended to improve isolated cognitive domains without aclear effect on overall cognition: minocycline for exampleimproved working memory, cognitive flexibility and planning(Levkovitz et al., 2010), galantamine improved verbalmemory and processing speed (Buchanan et al., 2008),while modafinil improved attentional set shifting in chronicpatients with schizophrenia (Turner et al., 2004) and work-ing memory in patients with first-episode psychosis (Scorielset al., 2012). The nicotinic alpha-7 agonist EVP-6124 hashowever demonstrated a significant effect on global cogni-tive function compared to placebo in patients with schizo-phrenia in a multi-site phase II trial in 319 patients withschizophrenia (News release of EnVivo pharmaceuticals athttp://www.envivopharma.com/news-item.php?id=35) andit is currently tested in larger phase III trials for CIAS inpatients with schizophrenia (Clinical trials gov. identifiersNCT01714661 and NCT01716975).

Finally, it is unclear if any of the cognitive improvementsinduced by cognitive-enhancing drugs are retained afterdrug discontinuation and whether they generalise to real-world functional improvement.

2.3. Rationale for combining cognitive-enhancingdrugs and cognitive remediation

Patients with schizophrenia have altered organisation andfunction of the neural circuits that support cognitiveprocesses (Minzenberg et al., 2009) as a consequenceof genetic vulnerabilities and environmental influencesthat shape this illness. While cognitive remediation hasbeen shown to induce neural changes in the cognitivecircuits, the associated therapeutic effect is modest. Theenhancement of the neural changes induced by cognitive

remediation with cognitive-enhancing drugs may result infurther improvements of the function of the disorderedcognitive circuits in schizophrenia and may lead to a level ofimprovement that neither intervention alone has achievedso far. The idea of combining a behavioural intervention(i.e. cognitive remediation) with cognitive-enhancing drugsas a therapeutic paradigm for CIAS has been suggested byseveral researchers in the field (e.g. Barch, 2010; Bowie andHarvey, 2006; Goff, 2012; Green, 2007; Swerdlow, 2011;Vinogradov et al., 2012).

A very illustrative example of the combination of acognitive behavioural intervention with a pharmacologicalcompound can be derived from preclinical evidence.Hampson et al. (1998) trained healthy laboratory rats on adelayed non-match to sample task where rats were requiredto press one of two bars, move to a new location wait therefor varying lengths of time, return to the test site and selectthe bar not previously pressed. After weeks of training ratsreached 77% of maximum performance and showed nofurther improvements with additional training. Every otherday injections with CX516, a positive modulator of AMPAglutamate receptors, prior to task performance, producedincreasing improvements in scores over a period of anadditional three weeks (Hampson et al., 1998). Theseresults suggest that a pharmacological compound caninduce enhancement of the cognitive effects of trainingwhere additional training alone does not. The combinationof training with a pharmacological compound in this casesurpassed what appeared to be a performance “ceiling” fortraining alone. Considering “the ceiling” of low-to-mediumrange cognitive improvements across a variety of cognitiveremediation interventions in patients with schizophrenia,a combination of cognitive remediation with cognitive-enhancing drugs could be a potentially effective way toenhance cognition even further and potentially lead togreater functional improvement.

The most studied pathway to enhancement of cognitiveperformance is through facilitation of learning and memory.The basic neurobiology of learning entails activity-depen-dent firing of particular neurons, accompanied by a sub-sequent, neurochemically mediated change in circuitstrength (Hebb, 1949). There is a wealth of evidence frompreclinical studies on drug-induced improvement of learningand memory dating back to 1917, when the first study onthe effects of strychnine on learning was published byLashley and where pre-training administration of strychninewas found to accelerate the rate of maze-learning in rats(Lashley, 1917). Preclinical studies investigating learningand memory in healthy or impaired laboratory animalstypically use one or more cognitive tasks and administerdrugs at different times and/or frequency either before orafter training on the cognitive task(s) in order to study theeffects of the drug on different aspects of learning andmemory (e.g. acquisition, storage, retention, retrieval,consolidation) (for reviews see Floresco and Jentsch, 2011and McGaugh and Roozendaal, 2009).

The principle of activity-dependent plasticity combinedwith pharmacological compounds has been applied for thepromotion of functional recovery after brain damage.Preclinical studies showed that a neuromodulatory pharma-cological compound (e.g. amphetamine) paired with circuit-specific physical training enhanced the recovery of function

793Treating impaired cognition in schizophrenia

in rodents and cats with brain (motor and visual) injuries(Feeney et al., 1982; Feeney and Hovda, 1985; Hovda andFenney, 1984).

Interestingly, the combination of activity-dependent plas-ticity with pharmacological compounds in laboratory ani-mals not only enhances the activity-dependent plasticeffects on the brain, as suggested by the above mentionedstudies, but can also reactivate plastic processes of earlierneurodevelopmental stages. Activity-dependent plasticchanges in the visual cortex of the rat are reduced after a“critical period” of postnatal development and monoculardeprivation in adult rats does not induce any changes in theocular dominance of visual cortex cells. However, whenmonocular deprivation in adult rats was paired with chronictreatment with the selective serotonin reuptake inhibitorfluoxetine, it induced domination of the visual cortex bymonocular cells responding to the stimulation of the openeye. Furthermore, plastic changes in the visual cortex of therat (i.e. shift in ocular dominance) during the “criticalperiod” of postnatal development induced by an “abnor-mal” activity-dependent plasticity model (i.e. monoculardeprivation) were reversed during adult life with thecombination of fluoxetine with a “symptom specific” activ-ity (i.e. shutting the previously open eye). Full restorationof a form of synaptic plasticity (White matter long-termpotentiation, WM-LTP) that is absent in the adults wasshown in the visual cortex of the fluoxetine-treated ratsonly. The reinstatement of neural plasticity in the visualcortex was also accompanied by increased BDNF proteinexpression in the visual cortex (Vetencourt et al., 2008).Recent preclinical data also suggest that the combination ofa pharmacological compound (fluoxetine) and a behavioural(extinction training) intervention resulted in behaviouralimprovements which neither intervention alone couldachieve in the adult rat. Fear extinction training in adultrats reduces fear responses temporarily but results inpermanent fear extinction in rats only during the “criticalperiod” of postnatal development. Fluoxetine treatment onits own does not eliminate fear responses in fear condi-tioned rats. However, the combination of fluoxetine andextinction training in adult rats resulted in long term fearextinction probably through reactivation of plastic changesof the “critical period” of postnatal development in theamygdala of the adult rats (Karpova et al., 2011).

In summary preclinical evidence suggests that a pharma-cological compound can augment and restore plasticity ofbehaviourally active neural circuits, which in turn mayresult in functional recovery.

In healthy humans, many studies have showed thatneuromodulatory drugs, such as amphetamine can enhancethe effects of motor training and result in faster develop-ment, increased magnitude, and longer lasting duration ofeffects compared to placebo (Bütefisch et al., 2002).Amphetamine (Breitenstein et al., 2004) as well as levodopaimproves learning of an artificial language across multipletraining sessions in healthy humans (Knecht et al., 2004).Modafinil, a wake-promoting agent, likely affects cognitionand learning via activation of dopaminergic, noradrenergicand histaminergic projections (Minzenberg and Carter, 2007)and has been shown to improve learning of an artificiallanguage across ten training sessions in healthy humans(unpublished data from our laboratory).

The ampakines are compounds that bind to a site on theAMPA-type glutamate receptor and enhance synapticresponses following the binding of glutamate to the recep-tor by stabilising the receptor in the channel-open state;they have been shown to improve learning and memory inpreclinical studies (Lynch and Gall, 2006). Ampakine CX516combined with training sessions in memory tasks in healthyvolunteers over 5 days improves performance in episodicand olfactory memory and also in incremental learning ofcomplex skills (Ingvar et al., 1997).

The combination of pharmacological compounds andsymptom-specific training has been used for post-strokerehabilitation. The effects of amphetamine (Crisostomoet al., 1988; Walker-Batson et al., 1995), methylphenidate(Grade et al., 1998), and levodopa (Scheidtmann et al.,2001) paired with motor rehabilitation for post-stroke motorrecovery have been investigated in small randomised con-trolled trials (RCTs). Amphetamine and methylphenidatecombined with motor rehabilitation accelerated the rate ofmotor recovery in hemiplegic stroke patients with large ESranging from Cohen's d=1.47–1.72 in the amphetaminestudies and Cohen's d=0.82 in the methylphenidate study.Amphetamine combined with speech/language rehabilita-tion also accelerated speech recovery in aphasic post-strokepatients with large ES (Cohen's d=1.19) (Walker-Batsonet al., 2001). In most cases the increased therapeutic gainswere retained after the discontinuation of the combinedintervention during follow-up, which ranged from 3 weeks(Scheidtmann et al., 2001) to 12 months (Walker-Batsonet al., 1995). While this preliminary evidence is encoura-ging, large double-blind, placebo-controlled trials are lack-ing and there are also negative results for the facilitatingeffects of D-amphetamine (Platz et al., 2005) and levodopa(Sonde and Lokk, 2007) in post-stroke motor recovery.Memantine, an NMDA “open channel blocker” that entersthe receptor channel and blocks current flow only afterchannel opening, without sharing the deleterious effects ofother compounds that also block NMDA receptor channels,such as ketamine (Johnson and Kotermanski, 2006), has alsobeen tested in combination with speech therapy in an RCTof28 patients with post-stroke aphasia for 2 weeks. Bothmemantine and speech therapy when administered aloneimproved aphasia, but their combination resulted in greaterimprovement of aphasia than either intervention on its own,with a large ES (Cohen's d=1.15) (Berthier et al., 2009).

In psychiatric therapeutics, the combination of D-cyclo-serine (DCS) with exposure behavioural therapy has beentested in patients with phobias and other anxiety disorders,based on studies of fear extinction in laboratory animals,which have shown that DCS, a partial agonist of the glycinesite of the N-methyl-D-aspartate receptor (NMDA) receptor,can accelerate the effects of fear extinction training.The studies in humans have shown that DCS paired withexposure therapy induces greater therapeutic gains relativeto exposure therapy alone, with a medium ES (Cohen'sd=0.6) and some decline of efficacy at follow-up (Cohen'sd=0.47). As DCS is behaviourally inactive on its own for fearextinction, it has been hypothesised that it increases theefficacy of exposure based psychotherapy by enhancingNMDA receptor function during fear extinction or by redu-cing NMDA receptor function during fear memory consolida-tion (Norberg et al., 2008).

P.G. Michalopoulou et al.794

The combination of antipsychotics with psychosocialinterventions and the resulting lower risk of relapse rates(e.g. Alvarez-Jiménez et al., 2011; Guo et al., 2010), bettercompliance and lower doses of antipsychotic medicationsand also greater gains from skill training provide an exampleof an analogous case of a combined intervention in schizo-phrenia (for review see Marder, 2000).

In summary, preclinical and clinical evidence suggeststhat the combination of a behavioural intervention with apharmacological compound can augment the effects ofthe behavioural intervention and may result in functionalrecovery.

Combining cognitive remediation with pharmacologicalcompounds leverages both these elements in the service ofpatients and has recently been termed as Pharmacologically-Augmented Cognitive Therapies (PACT) by Swerdlow (2012).We now turn our attention to the design issues and thepractical realities of PACT studies in schizophrenia.

2.4. Design issues and practical realities

Combining drugs with cognitive remediation would requireseveral issues to be addressed: Which drug? What type ofcognitive remediation? When to combine the two? What isthe outcome of interest?

2.5. Which cognitive-enhancing drugs?

Evidence from the preclinical and clinical studies sum-marised above suggests that there exist pharmacologicalcompounds that can modify the way neural circuits reorga-nise in response to training and such compounds could betested for their effects on cognitive remediation in patientswith schizophrenia. We will exemplify a few based on theevidence summarised above.

In theory, dopaminergic agents such as amphetamine andlevodopa could be considered for PACT studies in schizo-phrenia, as both enhance the response to training inpreclinical studies, studies in healthy individuals and clinicalpopulations summarised above. The chronic use of thesecompounds in patients with schizophrenia can be proble-matic due to their potential to exacerbate psychoticsymptoms. It is worth noting however that amphetaminehas been used safely in a few single-dose, cross-over studiesin chronic, stable, medicated patients with schizophrenia,where it improved cognitive performance (Barch and Carter,2005; Daniel et al., 1991; Goldberg et al., 1991; Pietrzaket al., 2010). In these studies no exacerbation of psychoticsymptoms was reported. Levodopa has also been used asadd-on in 5 placebo-controlled studies in patients withschizophrenia and a meta-analysis of these studies found abeneficial overall effect (because of the variety of outcomemeasures used in these studies, patients were divided intwo response groups on the basis of their response in thescale used in the individual studies for the purposes of themeta-analysis: poor response=no change or worsening andgood response= improved) with a medium-large ES (Cohen'sd=0.71) and no worsening of any symptoms of schizophre-nia (Jaskiw and Popli, 2004). There was however clinicalworsening in 38% of the patients who received high doses oflevodopa without concomitant antipsychotics. Based on the

above, low doses of such compounds in clinically stable,medicated patients with schizophrenia could potentiallyminimise the risk of exacerbation of psychotic symptomsand allow the investigation of their effects on cognitiveremediation in schizophrenia.

An associated, although not exclusively dopaminergic agent,modafinil, could also be considered as a candidate for PACTstudies in schizophrenia. Modafinil enhances the effects oftraining in an artificial language learning task in healthyvolunteers (unpublished data from our laboratory). Althoughmodafinil can be categorised as a psychostimulant due to itswake-promoting effects, it has neurochemical and behaviouraleffects distinct from those of amphetamine, including lowerabuse potential, lower cardiovascular side effects and possiblylower potential of exacerbation of psychotic symptoms com-pared to amphetamine (Minzenberg and Carter, 2007). Modafinilhas been used in 9 randomised trials in patients with schizo-phrenia or related conditions (see Scoriels et al., 2013 forreview) with only one report of a patient on modafinil whosepsychotic symptoms worsened during the first week of treat-ment (Sevy et al., 2005).

Among the glutamatergic compounds, DCS could be apotential candidate for PACT studies. Evidence from preclinicalstudies suggests that facilitation of memory consolidationenhances memory for many kinds of training experiences(Lupien and McEwen, 1997; Sandi and Rose, 1994) and thereis evidence from an RCT in patients with schizophrenia that DCSmay facilitate memory consolidation of newly learned material(Goff et al., 2008a). Following these findings, DSC was pairedwith training on cognitive-behavioural techniques for thefacilitation of generation of alternative explanations for thepatients' delusional beliefs in cross-over RCT in 20 patients withschizophrenia. There were no significant differences in thenumber of alternative explanations produced by patients onDCS compared to placebo, but patients who received DCS firstshowed greater reduction in the severity of delusions, asso-ciated distress and belief conviction compared to patients whoreceived placebo first. While the results of the study aredifficult to interpret, the authors suggest as a tentativeconclusion that DCS may have been associated with enhance-ment of learning only when following the first exposure totraining (Gottlieb et al., 2011).

Another glutamatergic compound, ampakine CX516,induced improvement of memory and learning in healthyhumans in combination with training in relevant tasks assummarised above (Ingvar et al., 1997) and did not produceany cognitive effects in patients with schizophrenia, whereit was however tested on its own, i.e. in the absence of anycognitive training (Goff et al., 2008b).

A final example of a drug acting at the glutamate NMDAreceptors that could be considered as candidate for PACTstudies in schizophrenia is memantine, which has beenshown to improve learning and memory and to enhancethe effects of speech training in post-stroke aphasicpatients. Memantine is currently being tested in an RCT incombination with cognitive training in patients with schizo-phrenia (NCT01555697).

It has been suggested that the choice of pharmacologicalcompounds for PACT studies could be guided by biomarkers,which would predict the ability of a compound to enhancethe effects of cognitive remediation (Chou et al., 2012).Prepulse inhibition of startle and electroencephalographic

795Treating impaired cognition in schizophrenia

measures, such as mismatch negativity and gamma bandsynchronisation, have been proposed as candidate biomar-kers on the basis of evidence suggesting that they arecorrelated with cognitive functions and/or certain geno-types (e.g. COMT genotypes), are impaired in patients withschizophrenia and may be sensitive to pharmacologicalcompounds. (Chou et al., 2012; Swerdlow 2011). Thepredictive value of such biomarkers (e.g. prepulse inhibitionfor memantine) is currently tested in PACT studies(NCT01555697).

2.6. Which type of cognitive remediation?

Given that CIAS spans almost all cognitive domains, rangingfrom the elementary, such as sensory-perceptual processes,to the highly complex, such as working memory, episodicmemory and executive functions (Dickinson et al., 2007),a question arises regarding the cognitive abilities thatshould be targeted by cognitive remediation. Evidence frompreclinical models of brain and spinal cord rehabilitationstress the importance of training on the specific behaviourthat needs to be enhanced, as non-specific rehabilitationtraining does not have any effect on specific behaviours(Fawcett, 2009). However, the guidance derived fromclinical studies in traumatic brain injury and stroke suggestsa broader approach (Cicerone et al., 2011). At present,cognitive remediation programmes that focus primarily onhigher order cognitive functions, either broadly or inisolation and also remediation programmes that exercisesimultaneously sensory as well as higher order cognitivefunctions have been tested in clinical trials in patients withschizophrenia. There is some evidence for cognitiveimprovement with all these different approaches but itremains unclear as to which aspects of these approaches arenecessary, sufficient or optimal (Wykes and Spaulding,2011). Recent meta-analytic evidence suggests that inaddition to focussing on specific cognitive domains, ele-ments of metacognition and coaching strategies combinedwith adjunctive psychosocial interventions may be criticalin transferring gains from a laboratory situation to the real-world context (Wykes et al., 2011).

2.7. When and how to combine cognitiveremediation with cognitive-enhancing drugs?

When in the course of illness one administers the drug andcognitive remediation combination may represent an impor-tant factor for optimal efficacy. There is a dearth ofevidence regarding the effects of illness duration and ageon the cognitive outcomes in clinical trials of potentialcognitive-enhancing drugs in schizophrenia. The majority ofthese trials have included patients in their 40 s with illnessduration of around 20 years (Keefe et al., 2011) and nostudies so far have compared the effects of a cognitive-enhancing drug in patients with schizophrenia with differentduration of illness.

Similarly, the majority of cognitive remediation studies sofar have also been conducted in patients with chronicschizophrenia with positive, although modestly so, results.Recent preclinical findings in a neurodevelopmental schizo-phrenia model using neonatal ventral hippocampus lesion

rats, showed that cognitive training during adolescenceprevented the adult cognitive impairment associated withthe hippocampal lesion (Lee et al., 2012); and cognitiveremediation in the early course of schizophrenia has beenassociated with neuroprotective effects (Eack et al., 2010).Meta-analytical evidence however suggests that despiteindividual study evidence, the age of participants does nothave an effect on the cognitive outcomes of cognitiveremediation in schizophrenia (Wykes et al., 2011).

Overall, there is currently no available evidence tosuggest a definite temporal window of opportunity fortreating cognitive impairment in schizophrenia and the roleof illness duration and age in the outcome of PACT studiesshould be addressed in future studies.

The optimal duration of the intervention needs also to beaddressed. The average length of treatment in cognitiveremediation studies was 32.2 h (range=4–130), providedacross 16.7 weeks (range=2–104). However, the durationof the cognitive remediation programmes did not have asignificant effect on cognitive change (Wykes et al., 2011).

An associated issue is that of maintenance therapy.Cognitive remediation programmes require more time andeffort than taking a drug or a psychological treatmentalone, and the relevant question is whether a singlemodality therapy would maintain the gains. There is noavailable evidence so far regarding maintenance therapyfrom the studies that have used combined treatments inpost-stroke rehabilitation and in anxiety disorders and thisissue remains open for future successful PACT studies.

Since no PACT studies are currently available, what isneeded is a stepped strategy. In its first instance, one canonly seek evidence of PACT under laboratory settings usingdiscrete cognitive tasks to optimise issues of drug choice,drug dose and choice of cognitive remediation programme.The first studies in schizophrenia will provide the proof-of-concept, i.e. the demonstration that PACT can result inimprovement in some aspects of cognitive function that islarger than that associated with either intervention andremains present after the cessation of the combined inter-vention. Only once that is established will it be possible toundertake larger real-world studies, where the effects ofparticipants' characteristics (e.g. illness duration, age,baseline cognitive impairment, type and dose of antipsy-chotic medication), the duration of the intervention, andthe maintenance therapy on the cognitive outcomes of PACTstudies will be investigated systematically.

A challenge unique to PACT studies is exactly when thecognitive-enhancing drug should be administered relative tocognitive remediation. All drugs show changing plasmalevels through the day and ideally cognitive remediationshould coincide with the time window of maximal plasticityenhancement by the cognitive-enhancing drug. This isparticularly critical as the point of PACT is to specificallyenhance cognitive remediation-related neural plasticchanges. At this point there are no independent measuresof the plasticity window and therefore drug plasma levelsserve as the most relevant proxy; it is thus customary inpreclinical and clinical studies to pair the training with thepeak plasma levels of the drug.

A final challenge for PACT is the outcome measure bywhich its success should be assessed, and if successful, howit might be implemented in the clinic. The ultimate goal of

P.G. Michalopoulou et al.796

addressing CIAS is to improve real-world functioning. Therelationship between cognition and function in schizophre-nia is robust but the way cognitive improvements changefunctioning is still an open question (Wykes et al., 2012).Nevertheless, meta-analytic evidence suggests that strongereffects on function are produced when cognitive remediationis added to other rehabilitation programs (Wykes et al., 2011).Recent studies (Bowie et al., 2012) have found that thecombination of cognitive remediation with functional skillstraining enhances the effects of cognitive remediation ineveryday functioning. It is therefore likely that translation ofany cognitive benefits from PACT studies to real-world func-tioning may require additional functioning-oriented interven-tions (e.g. supported employment or bridging/generalisationgroups).

3. Practical issues

Clearly a combined treatment such as above is morecomplex to develop and test than a single modality treat-ment. Pursuing PACT will have significant implications forthe pivotal trials, the approval process by the regulatoryagencies as well as subsequent marketing.

First, at the pivotal trial stage one would need a doubledummy design (i.e. a control for the cognitive-enhancingdrug – placebo and a control for cognitive remediation –

some form of non-cognitive supportive therapy), whichwould increase the number of arms in the study from thecustomary two to four. Second, while sponsors routinelystandardise medication administration, such trials wouldrequire sponsors to additionally validate and providemulticentre cognitive remediation. There is ongoing pro-gress in this respect from the Cognitive RemediationExperts Workshop (CREW) meeting in Florence in 2012,where the definition of the factors that make up aneffective cognitive remediation programme was discussedby the cognitive remediation experts group; this definitioncould drive future PACT studies.

Regarding the approval process by the regulatory agen-cies, there are examples such as varenicline, a partialagonist selective for α4β2 nicotinic acetylcholine receptors,for smoking cessation, which is approved by the FDA andrecommended by the Agency for Healthcare Research andQuality in combination with counselling to support theattempt to cease smoking. However, this indication camein the context of clinical trials where background counsel-ling was the baseline treatment-as-usual for smoking cessa-tion. In the case of PACT studies, cognitive remediation isnot as yet an approved treatment-as-usual for CIAS inschizophrenia. Therefore, PACT studies in schizophreniamight require independent arms testing the cognitive-enhancing drug, the cognitive remediation approach and acombination of the two with appropriate dummies andcontrols. All these challenging practical issues associatedwith PACT studies in schizophrenia have hindered theemployment of this strategy in more studies so far. No suchbundled drug-therapy product has as yet been approved bythe regulatory agencies and given the complexity of issuesto be resolved many companies may shy away from seekingformal approval for such an approach.

4. Conclusion

Addressing CIAS is the next frontier in schizophrenia. Thereare currently no indicated cognitive-enhancing drugs and nowidely accepted or applied cognitive remediation approach.We provide the theoretical rationale, the preliminaryevidence and the design and practical challenges that besetthis approach. There are several unanswered questionsabout how one would use such a strategy and it is certainlymore complex than administering either intervention alone.However, since CIAS has not as yet effectively yielded tocognitive-enhancing drugs and only modestly to cognitiveremediation, the field is urged to explore their combination.

Role of funding source

This work was supported by the Innovative Medicines Initiative JointUndertaking—Novel Methods leading to New Medications in Depres-sion and Schizophrenia (NEWMEDS) (Grant Agreement no. 115008), ofwhich resources are composed of EFPIA in-kind contribution andfinancial contribution from the European Union's Seventh FrameworkProgramme (FP7/2007–2013) and by Medical Research Council (UK)Strategic Appointments Scheme (Grant no. G0701748) to Prof. Kapur.

Contributors

PGM, SWL, TW, JG and SK discussed the original idea of the review.PGM performed literature search and wrote all the versions ofthe review manuscript. SK provided supervision and guidance in thereview writing. SWL, TW, JG and SK contributed to editing of all themanuscript versions. All the authors contributed to and haveapproved the final manuscript.

Conflict of interest

Prof. Kapur has received grant support from AstraZeneca, Bristol-Myers Squibb and GlaxoSmthKlline and has served as consultant,scientific advisor and had speaking engagements for AstraZeneca,Bioline, Bristol Meyers Squibb, Eli Lilly, Janssen (Johnson andJohnson), Lundbeck, NeuroSearch, Otsuka, Pfizer, Roche, Servier,Solvay and Wyeth.

At the time this work was conducted Dr Jaeger was an employee ofand shareholder in AstraZeneca Pharmaceuticals and was aware of noconflicts of interest with respect to the contents of this article.Dr Jaeger is currently an employee of CogState and is aware of noconflicts of interest with respect to the contents of this article.

Dr Michalopoulou, Prof. Lewis and Prof. Wykes declare no conflictof interest.

Acknowledgements

Prof. Wykes and Prof. Kapur acknowledge the support provided by NIHRBiomedical Research Centre for Mental Health at the South London andMaudsley NHS Foundation Trust and Institute of Psychiatry, King's CollegeLondon. The views expressed are those of the authors and notnecessarily those of the NHS, the NIHR or the Department of Health.

References

Alvarez-Jiménez, M., Parker, A.G., Hetrick, S.E., McGorry, P.D.,Gleeson, J.F., 2011. Preventing the second episode: a systematicreview and meta-analysis of psychosocial and pharmacologicaltrials in first-episode psychosis. Schizophr. Bull. 37, 619–630.

797Treating impaired cognition in schizophrenia

Barch, D.M., 2010. Pharmacological strategies for enhancing cogni-tion in schizophrenia. Curr. Top. Behav. Neurosci. 4, 43–96.

Barch, D.M., Carter, C.S., 2005. Amphetamine improves cognitivefunction in medicated individuals with schizophrenia and inhealthy volunteers. Schizophr. Res. 77, 43–58.

Berthier, M.L., Green, C., Lara, J.P., Higueras, C., Barbancho, M.A.,Dávila, G., Pulvermüller, F., 2009. Memantine and constraint-induced aphasia therapy in chronic poststroke aphasia. Ann.Neurol. 65, 577–585.

Bowie, C.R., Harvey, P.D., 2006. Treatment of cognitive deficits inschizophrenia. Curr. Opin. Investig. Drugs 7, 608–613.

Bowie, C.R., McGurk, S.R., Mausbach, B., Patterson, T.L., Harvey, P.D.,2012. Combined cognitive remediation and functional skills trainingfor schizophrenia: effects on cognition, functional competence, andreal-world behavior. Am. J. Psychiatry 169, 710–718.

Breitenstein, C., Wailke, S., Bushuven, S., Kamping, S., Zwitserlood, P.,Ringelstein, E.B., Knecht, S., 2004. D-amphetamine boosts languagelearning independent of its cardiovascular and motor arousingeffects. Neuropsychopharmacology 29, 1704–1714.

Buchanan, R.W., Conley, R.R., Dickinson, D., Ball, M.P., Feldman, S.,Gold, J.M., McMahon, R.P., 2008. Galantamine for the treatment ofcognitive impairments in people with schizophrenia. Am. J.Psychiatry 165, 82–89.

Buchanan, R.W., Freedman, R., Javitt, D.C., Abi-Dargham, A.,Lieberman, J.A., 2007. Recent advances in the development ofnovel pharmacological agents for the treatment of cognitiveimpairments in schizophrenia. Schizophr. Bull. 33, 1120–1130.

Bütefisch, C.M., Davis, B.C., Sawaki, L., Waldvogel, D., Classen, J.,Kopylev, L., Cohen, L.G., 2002. Modulation of use-dependentplasticity by D-amphetamine. Ann. Neurol. 51, 59–68.

Chou, H.H., Twamley, E., Swerdlow, N.R., 2012. Towardsmedication-enhancement of cognitive interventions in schizo-phrenia. Handb. Exp. Phaermacol. 213, 81–111.

Cicerone, K.D., Langenbahn, D.M., Braden, C., Malec, J.F., Kalmar, K.,Fraas, M., Felicetti, T., Laatsch, L., Harley, J.P., Bergquist, T.,Azulay, J., Cantor, J., Ashman, T., 2011. Evidence-based cognitiverehabilitation: updated review of the literature from 2003 through2008. Arch. Phys. Med. Rehabil. 92, 519–530.

Crisostomo, E.A., Duncan, P.W., Propst, M., Dawson, D.V., Davis, J.N.,1988. Evidence that amphetamine with physical therapy promotesrecovery of motor function in stroke patients. Ann. Neurol. 23,94–97.

Daniel, D.G., Weinberger, D.R., Jones, D.W., Zigun, J.R., Coppola, R.,Handel, S., Bigelow, L.B., Goldberg, T.E., Berman, K.F., Kleinman,J.E., 1991. The effect of amphetamine on regional cerebral bloodflow during cognitive activation in schizophrenia. J. Neurosci. 11,1907–1917.

Dickinson, D., Ramsey, M.E., Gold, J.M., 2007. Overlooking theObvious: a meta-analytic comparison of digit symbol codingtasks and other cognitive measures in schizophrenia. Arch. Gen.Psychiatry 64, 532–542.

Dickinson, D., Tenhula, W., Morris, S., Brown, C., Peer, J., Spencer, K.,Li, L., Gold, J.M., Bellack, A.S., 2010. A randomized, controlled trialof computer-assisted cognitive remediation for schizophrenia. Am.J. Psychiatry 167, 170–180.

Dixon, L.B., Dickerson, F., Bellack, A.S., Bennett, M., Dickinson, D.,Goldberg, R.W., Lehman, A., Tenhula, W.N., Calmes, C., Pasil-las, R.M., Peer, J., Kreyenbuhl, J., 2010. The 2009 schizophreniaPORT psychosocial treatment recommendations and summarystatements. Schizophr. Bull. 36, 48–70.

Eack, S.M., Hogarty, G.E., Cho, R.Y., Prasad, K.M.R., Greenwald, D.P.,Hogarty, S.S., Keshavan, M.S., 2010. Neuroprotective effects ofcognitive enhancement therapy against gray matter loss in earlyschizophrenia: results from a 2-year randomized controlled trial.Arch. Gen. Psychiatry 67, 674–682.

Ehrenreich, H., Hinze-Selch, D., Stawicki, S., Aust, C.,Knolle-Veentjer, S., Wilms, S., Heinz, G., Erdag, S., Jahn, H.,Degner, D., Ritzen, M., Mohr, A., Wagner, M., Schneider, U.,

Bohn, A.L., Huber, M., Czernik, A., Pollmacher, T., Maier, W.,Siren, A.L., Klosterkotter, J., Falkai, P., Ruther, E., Aldenhoff, J.B., Krampe, H., 2007. Improvement of cognitive functions inchronic schizophrenic patients by recombinant human erythro-poietin. Mol. Psychiatry 12, 206–220.

Fawcett, J., 2009. Molecular control of brain plasticity and repair.Prog. Brain Res. 175, 501–509.

Feeney, D.M., Gonzalez, A., Law, W.A., 1982. Amphetamine,haloperidol, and experience interact to affect rate of recoveryafter motor cortex injury. Science 217, 855–857.

Feeney, D.M., Hovda, D.A., 1985. Reinstatement of binocular depthperception by amphetamine and visual experience after visualcortex ablation. Brain Res. 342, 352–356.

Floresco, S.B., Jentsch, J.D., 2011. Pharmacological enhancementof memory and executive functioning in laboratory animals.Neuropsychopharmacology 36, 227–250.

Goff, D.C., Cather, C., Gottlieb, J.D., Evins, A.E., Walsh, J., Raeke, L.,Otto, M.W., Schoenfeld, D., Green, M.F., 2008a. Once-weeklyD-cycloserine effects on negative symptoms and cognition in schizo-phrenia: an exploratory study. Schizophr. Res. 106, 320–327.

Goff, D.C., Lamberti, J.S., Leon, A.C., Green, M.F., Miller, A.L.,Patel, J., Manschreck, T., Freudenreich, O., Johnson, S.A.,2008b. A placebo-controlled add-on trial of the Ampakine,CX516, for cognitive deficits in schizophrenia. Neuropsychophar-macology 33, 465–472.

Goff, D.C., 2012. Memory consolidation deficits in schizophreniaand the combination of D-cycloserine with cognitive remedia-tion. Neuropsychopharmacology 38, S21.

Goldberg, T.E., Bigelow, L.B., Weinberger, D.R., Daniel, D.G.,Kleinman, J.E., 1991. Cognitive and behavioral effects of thecoadministration of dextroamphetamine and haloperidol inschizophrenia. Am. J. Psychiatry 148, 78–84.

Gottlieb, J.D., Cather, C., Shanahan, M., Creedon, T., Macklin, E.A.,Goff, D.C., 2011. D-cycloserine facilitation of cognitive behavioraltherapy for delusions in schizophrenia. Schizophr. Res. 131, 69–74.

Grade, C., Redford, B., Chrostowski, J., Toussaint, L., Blackwell, B.,1998. Methylphenidate in early poststroke recovery:a double-blind, placebo-controlled study. Arch. Phys. Med. Reha-bil. 79, 1047–1050.

Green, M.F., Kern, R.S., Braff, D.L., Mintz, J., 2000. Neurocognitivedeficits and functional outcome in schizophrenia: are wemeasuring the “right stuff”? Schizophr. Bull. 26, 119–136.

Green, M.F., 2007. Stimulating the development of drug treatmentsto improve cognition in schizophrenia. Annu. Rev. Clin. Psychol.3, 159–180.

Guo, X., Zhai, J., Liu, Z., Fang, M., Wang, B., Wang, C., Hu, B., Sun, X.,Lv, L., Lu, Z., Ma, C., He, X., Guo, T., Xie, S., Wu, R., Xue, Z.,Chen, J., Twamley, E.W., Jin, H., Zhao, J., 2010. Effect ofantipsychotic medication alone vs combined with psychosocialintervention on outcomes of early-stage schizophrenia: a rando-mized, 1-year study. Arch. Gen. Psychiatry 67, 895–904.

Hampson, R.E., Rogers, G., Lynch, G., Deadwyler, S.A., 1998.Facilitative effects of the ampakine CX516 on short-termmemory in rats: enhancement of delayed-nonmatch-to-sampleperformance. J. Neurosci. 18, 2740–2747.

Harvey, P.D., Howanitz, E., Parrella, M., White, L., Davidson, M.,Mohs, R.C., Hoblyn, J., Davis, K.L., 1998. Symptoms, cognitivefunctioning, and adaptive skills in geriatric patients with lifelongschizophrenia: a comparison across treatment sites. Am. J.Psychiatry 155, 1080–1086.

Hebb, D.O., 1949. The Organization of Behavior: A Neuropsycholo-gical Review. Wiley, New York.

Hovda, D.A., Fenney, D.M., 1984. Amphetamine with experiencepromotes recovery of locomotor function after unilateral frontalcortex injury in the cat. Brain Res. 298, 358–361.

Ingvar, M., Ambros-Ingerson, J., Davis, M., Granger, R., Kessler, M.,Rogers, G.A., Schehr, R.S., Lynch, G., 1997. Enhancement by an

P.G. Michalopoulou et al.798

ampakine of memory encoding in humans. Exp. Neurol. 146,553–559.

Jaskiw, G., Popli, A., 2004. A meta-analysis of the response tochronic l-dopa in patients with schizophrenia: therapeutic andheuristic implications. Psychopharmacology (Berl.) 171,365–374.

Johnson, J.W., Kotermanski, S.E., 2006. Mechanism of action ofmemantine. Curr. Opin. Pharmacol. 6, 61–67.

Karpova, N.N., Pickenhagen, A., Lindholm, J., Tiraboschi, E.,Kulesskaya, N., Ãgustsdottir, A., Antila, H., Popova, D., Aka-mine, Y., Sullivan, R., Hen, R., Drew, L.J., Castren, E., 2011.Fear erasure in mice requires synergy between antidepressantdrugs and extinction training. Science 334, 1731–1734.

Keefe, R.S.E., Buchanan, R.W., Marder, S.R., Schooler, N.R., Dugar, A.,Zivkov, M., Stewart, M., 2011. Clinical trials of potential cognitive-enhancing drugs in schizophrenia: what have we learned so far?Schizophr. Bull.http://dx.doi.org/10.1093/schbul/sbr153.

Knecht, S., Breitenstein, C., Bushuven, S., Wailke, S., Kamping, S.,Floel, A., Zwitserlood, P., Ringelstein, E.B., 2004. Levodopa:faster and better word learning in normal humans. Ann. Neurol.56, 20–26.

Lashley, K.S., 1917. The effects of strychnine and caffeine upon therate of learning. Psychobiology 1, 141–169.

Lee, H., Dvorak, D., Kao, H.Y., Duffy, A.M., Scharfman, H.E.,Fenton, A.A., 2012. Early cognitive experience prevents adultdeficits in a neurodevelopmental schizophrenia model. Neuron75, 714–724.

Levkovitz, Y., Mendlovich, S., Riwkes, S., Braw, Y., Levkovitch-Verbin, H., Gal, G., Fennig, S., Treves, I., Kron, S., 2010.A double-blind, randomized study of minocycline for the treat-ment of negative and cognitive symptoms in early-phase schizo-phrenia. J. Clin. Psychiatry 71, 138–149.

Lupien, S.J., McEwen, B.S., 1997. The acute effects of corticoster-oids on cognition: integration of animal and human modelstudies. Brain Res. Brain Res. Rev. 24, 1–27.

Lynch, G., Gall, C.M., 2006. Ampakines and the threefold path tocognitive enhancement. Trends Neurosci. 29, 554–562.

Marder, S.R., 2000. Integrating pharmacological and psychosocialtreatments for schizophrenia. Acta Psychiatr. Scand. Suppl., 87–90.

McGaugh, J.L., Roozendaal, B., 2009. Drug enhancement of mem-ory consolidation: historical perspective and neurobiologicalimplications. Psychopharmacology (Berl.) 202, 3–14.

Minzenberg, M.J., Carter, C.S., 2007. Modafinil: a review ofneurochemical actions and effects on cognition. Neuropsycho-pharmacology 33, 1477–1502.

Minzenberg, M.J., Laird, A.R., Thelen, S., Carter, C.S., Glahn, D.C.,2009. Meta-analysis of 41 functional neuroimaging studies ofexecutive function in schizophrenia. Arch. Gen. Psychiatry 66,811–822.

National Institute for Health and Clinical Excellence (NICE) 2009.National Collaborating Centre for Mental Health: Schizophrenia:Core interventions in the treatment and management of schizo-phrenia in the primary and secondary care. London.

Norberg, M.M., Krystal, J.H., Tolin, D.F., 2008. A meta-analysis of D-Cycloserine and the facilitation of fear extinction and exposuretherapy. Biol. Psychiatry 63, 1118–1126.

Pietrzak, R.H., Snyder, P.J., Maruff, P., 2010. Use of an acutechallenge with D-amphetamine to model cognitive improve-ment in chronic schizophrenia. Hum. Psychopharmacol. 25, 353–358.

Platz, T., Kim, I.H., Engel, U., Pinkowski, C., Eickhof, C., Kutzner,M., 2005. Amphetamine fails to facilitate motor performanceand to enhance motor recovery among stroke patients with mildarm paresis: interim analysis and termination of a double blind,randomised, placebo-controlled trial. Restor. Neurol. Neurosci.23, 271–280.

Reichenberg, A., Harvey, P.D., 2007. Neuropsychological impair-ments in schizophrenia: integration of performance-based andbrain imaging findings. Psychol. Bull. 133, 833–858.

Sandi, C., Rose, S.P., 1994. Corticosterone enhances long-termretention in one-day-old chicks trained in a weak passiveavoidance learning paradigm. Brain Res. 647, 106–112.

Scheidtmann, K., Fries, W., Müller, F., Koenig, E., 2001. Effect oflevodopa in combination with physiotherapy on functional motorrecovery after stroke: a prospective, randomised, double-blindstudy. Lancet 358, 787–790.

Scoriels, L., Barnett, J., Soma, P., Sahakian, B., Jones, P., 2012.Effects of modafinil on cognitive functions in first episodepsychosis. Psychopharmacology (Berl.) 220, 249–258.

Scoriels, L., Jones, P.B., Sahakian, B.J., 2013. Modafinil effects oncognition and emotion in schizophrenia and its neurochemicalmodulation in the brain. Neuropharmacology 64, 168–184.

Sevy, S., Rosenthal, M.H., Alvir, J., Meyer, S., Visweswaraiah, H.,Gunduz-Bruce, H., Schooler, N.R., 2005. Double-blind, placebo-controlled study of modafinil for fatigue and cognition inschizophrenia patients treated with psychotropic medications.J. Clin. Psychiatry 66, 839–843.

Sonde, L., Lokk, J., 2007. Effects of amphetamine and/or L-dopaand physiotherapy after stroke—a blinded randomized study.Acta Neurol. Scand. 115, 55–59.

Swerdlow, N.R., 2011. Are we studying and treating schizophreniacorrectly? Schizophr. Res. 130, 1–10.

Swerdlow, N.R., 2012. Beyond antipsychotics: pharmacologically-augmented cognitive therapies (PACTs) for schizophrenia. Neu-ropsychopharmacology 37, 310–311.

Turner, D.C., Clark, L., Pomarol-Clotet, E., McKenna, P., Robbins, T.W.,Sahakian, B.J., 2004. Modafinil improves cognition and attentionalset shifting in patients with chronic schizophrenia. Neuropsycho-pharmacology 29, 1363–1373.

Vetencourt, J.F.M., Sale, A., Viegi, A., Baroncelli, L., De Pasquale,R., O'Leary, F., Castren, O., Maffei, L., E., 2008. The antide-pressant fluoxetine restores plasticity in the adult visual cortex.Science 320, 385–388.

Vinogradov, S., Fisher, M., Holland, C., Shelly, W., Wolkowitz, O.,Mellon, S.H., 2009. Is serum brain-derived neurotrophic factor abiomarker for cognitive enhancement in schizophrenia? Biol.Psychiatry 66, 549–553.

Vinogradov, S., Fisher, M., de Villers-Sidani, E., 2012. Cognitivetraining for impaired neural systems in neuropsychiatric illness.Neuropsychopharmacology 37, 43–76.

Walker-Batson, D., Curtis, S., Natarajan, R., Ford, J., Dronkers, N.,Salmeron, E., Lai, J., Unwin, D.H., 2001. A double-blind,placebo-controlled study of the use of amphetamine in thetreatment of aphasia. Stroke 32, 2093–2098.

Walker-Batson, D., Smith, P., Curtis, S., Unwin, H., Greenlee, R.,1995. Amphetamine paired with physical therapy acceleratesmotor recovery after stroke: further evidence. Stroke 26,2254–2259.

Wykes, T., Brammer, M., Mellers, J., Bray, P., Reeder, C., Williams, C.,Corner, J., 2002. Effects on the brain of a psychological treatment:cognitive remediation therapy: functional magnetic resonanceimaging in schizophrenia. Br. J. Psychiatry 181, 144–152.

Wykes, T., Huddy, V., Cellard, C., McGurk, S.R., Czobor, P., 2011.A meta-analysis of cognitive remediation for schizophrenia:methodology and effect sizes. Am. J. Psychiatry 168, 472–485.

Wykes, T., Spaulding, W.D., 2011. Thinking about the futurecognitive remediation therapy—what works and could we dobetter? Schizophr. Bull. 37, S80–S90.

Wykes, T., Reeder, C., Huddy, V., Taylor, R., Wood, H., Chirasim,N., Kontis, D., Landau, S., 2012. Developing models of howcognitive impairments change functioning: mediation, modera-tion and moderated mediation. Schizophr. Res. 138, 88–93.