CognitiveDrugResearchLtd.,ReadingRG301EA,UKMany people chew gum partly due to the belief that itincreases aspects of mental performance, including concen-tration. To the best of our knowledge no empirical evidenceexists to support this contention. The present experiment,therefore, examined the effects of chewing gum using acomprehensive and sensitive cognitive assessment batteryand two tasks manipulating cognitive load. Heart rateresponses were also measured.Seventy-five healthy adult participants (mean age 246years) were randomly assigned to one of three experimentalconditions (N‹25 per group): ‘‘chewing’’ – a piece of sugar-free chewing gum (Wrigley’s Extra Spearmint) was chewednaturally and constantly throughout the procedure; ‘‘shamchewing’’ – participants mimicked chewing movements in theabsence of gum; ‘‘quiet control’’ – no chewing behaviour wasperformed.Aspects of attention, working memory and long-termmemory were assessed using the Cognitive Drug Research(CDR) computerised battery. Stimuli were presented on acolour monitor and, except for two written word recall tasks,responses were collected automatically using a ‘‘Yes’’/‘‘No’’responsemodule.Thetaskswerepresentedintheorder:Wordpresentation, Immediate Word Recall, Picture Presentation,SimpleReactionTime,DigitVigilance,ChoiceReactionTime,Spatial Working Memory, Numeric Working Memory,Delayed Word Recall, Word Recognition and PictureRecognition (for details, e.g. see Kennedy et al. 2000).Following the CDR battery, participants performed compu-terised Serial Subtractions tasks. These assess concentrationand working memory and allow manipulation of cognitiveload (see Scholey et al., 2001 for details). In the present studySerialThrees(involvingtherepeatedsubtractionofthreefroma randomly generated starting number using the computer’snumerickeypad)thenSerialSevens(subtractionofseven)wereused, each for 2min.Eachcognitive taskoutcome measurewasanalysedby one-wayanalysisofvariance,withDunnettcomparisonstoisolatebetween-group effects where appropriate. The most strikingfindingwasasignificanteffectonbothimmediateanddelayedword recall, with more words being recalled in the chewingconditioncomparedwiththequietcontrolcondition(Table1).The Spatial Working Memory sensitivity index and Numericworkingmemoryreactiontimeweresimilarlyimprovedinthechewingcondition,andalsointheshamchewingconditionforthe latter measure (which reflects the efficiency of workingmemory operations). In addition, simple reaction times wereslowerintheshamchewingconditionthaninthequietcontrolcondition.Baseline heart rate recordings (sampled at 30-s intervals)began 240s prior to treatment and continued during a 180-speriod of chewing, sham chewing or sitting quietly prior tocognitive assessment (which lasted about 30min in all). Heartrate (mean bpm) was calculated during baseline, treatment,each of the 10 CDR tasks and both Serial Subtraction tasks.Heart rate changes relative to baseline were subjected to a3(Condition) 14(Phase) factorial ANOVA with repeatedmeasures on the latter factor. The main effect of conditionapproached significance, F(2,936)‹30, p‹006; heart ratesweresignificantlyhigherinthechewingconditionthaninquietcontrols, p<005 (Fig. 1). There was also a significant maineffect oftask,F(13,936)‹130,p<001:with theexceptionofSimple Reaction Time and Delayed Word Recall, all taskphasesofthestudywereassociatedwithsignificantincreasesinheart rate. There was also a significant task conditioninteraction, F(26,936)‹256, p<001.Theseresultsprovidedthefirstevidencethatthechewingofgum can improve episodic memory (involving the learning,storage and retrieval of information) and working memory(where information is held ‘‘on line’’). They did not indicatethat gum-chewing improves aspects of attention, at least asmeasured here.The impaired Simple Reaction Time during sham chewingmay reflect diversion of attentional resources during initialstages of performing this unfamiliar behaviour. This is con-sistent with the elevated heart rate observed while shamchewing in earlierphases of the experiment (Fig. 1). Althoughanactivecontrolisimportant,shamchewingmaynotbeidealfor this purpose, because most cognitive scores (except, nota-E-mail: a.scholey@unn.ac.uk bly, Numeric Working Memory Reaction Time) in this group0195–6663/02/$ – see front matter # 2002 Elsevier Science Ltd. All rights reserved.