Research report

Exposure to foodsrsquo non-taste sensory properties A nurseryintervention to increase childrenrsquos willingness to try fruitand vegetables

Paul Dazeley a Carmel Houston-Price ba School of Psychology amp Clinical Language Sciences University of Reading Earley Gate Whiteknights Reading RG6 6AL UKb School of Psychology amp Clinical Language Sciences University of Reading Malaysia Menara Kotaraya Level 7 Jalan Trus 80000 Johor Bahru Malaysia

A R T I C L E I N F O

Article historyReceived 2 June 2014Received in revised form 14 August 2014Accepted 29 August 2014Available online 10 September 2014

KeywordsRepeated exposureSensory interactionFruit and vegetablesNursery interventionFood familiarity

A B S T R A C T

Activities that engage young children with the sensory properties of foods are popular with nursery schoolsdespite the lack of evidence for their efficacy in increasing childrenrsquos consumption of healthy foods Thisstudy provides the first empirical exploration of the effectiveness of a non-taste sensory activity programin a nursery school setting Ninety-two children aged between 12 and 36 months were allocated eitherto an intervention group who took part in looking listening feeling and smelling activities with unusualfruits and vegetables every day for 4 weeks or to a non-intervention control group In a subsequent meal-time taste test children touched and tasted more of the vegetables to which they had been familiarizedin their playtime activities than of a matched set of non-exposed foods The results demonstrate thathands-on activities with unfamiliar fruits and vegetables can enhance childrenrsquos willingness to taste thesefoods and confirm the potential for such activities to support healthy eating initiatives

copy 2014 The Authors Published by Elsevier Ltd This is an open access article under the CC BY license(httpcreativecommonsorglicensesby30)

Introduction

Infantsrsquo taste preferences are initially determined by innate pre-dispositions such as a liking for sweet tastes (Desor Maller amp Andrews1975 Desor Maller amp Turner 1973) which is assumed to providean evolutionary benefit by identifying calorific foods (Wardle amp Cooke2008) Additionally children commonly demonstrate a reluctance totaste unfamiliar foods as they approach the age of 2 years this lsquofoodneophobiarsquo (Cooke 2007 Rozin 1976) is thought to protect the childfrom ingesting potentially harmful substances as they become ableto explore their environment more independently Both these in-stinctive behavioral patterns decrease the likelihood that children willdevelop a liking of a wide variety of healthy foods particularly veg-etables which often have a bitter taste

Considerable work has shown that familiarizing young chil-dren with the taste of an unfamiliar or disliked food is an effectivemeans of increasing their acceptance of it (Birch 1999) between10 and 15 taste exposures are generally found to be required (Birchamp Marlin 1982 Gerrish amp Mennella 2001 Wardle et al 2003a

Wardle Herrera Cooke amp Gibson 2003b) While it is often assumedthat the active component of such exposure interventions is the re-peated tasting of the food and the lsquolearned safetyrsquo that results fromthis (Kalat amp Rozin 1973 Rozin 1976) the act of tasting is not anisolated sensory experience as it also provides exposure to the foodrsquosnon-taste sensory qualities That is when a food is offered to a childto eat the child is exposed to the sight of the food its smell andits texture as the food is handled he or she may also hear the nameof the food If they taste it they additionally experience the textureof the food in the mouth and the sound it makes as it is chewedFamiliarity with these non-taste sensory properties of a food mayplay some part in the positive effects that result fromrepeated tasting

Research suggests that despite its evidential success repeat-edly offering children foods to taste is not a technique widely usedby parents at home 80 of caregivers are only prepared to offer theirchild a new food three or four times before they will decide thattheir child does not like it on the basis of the childrsquos lsquobothersomebehaviorrsquo (Carruth Ziegler Gordon amp Barr 2004) Parents may findit easier to provide repeated exposure to the non-taste propertiesof foods especially if this occurs outside mealtimes in an environ-ment that does not carry the stress associated with ensuring thatthe child is consuming a healthy diet There is therefore value inexploring the potential for non-taste familiarization to facilitate theintroduction of new foods

Nursery schools in the UK are increasingly signing up to multi-sensory activity programs that seek to encourage healthy eating by

Acknowledgements This project was completed as part of a Knowledge Trans-fer Partnership (Ref KTP007782) between the University of Reading and Ellarsquos Kitchen(Brands) Ltd The project was awarded funding by Innovate UK the UKrsquos innova-tion agency with the ESRC as the main funder The authors experienced no conflictsof interest during the preparation of the manuscript

Corresponding authorE-mail address chouston-pricereadingacuk (C Houston-Price)

httpdxdoiorg101016jappet2014080400195-6663copy 2014 The Authors Published by Elsevier Ltd This is an open access article under the CC BY license (httpcreativecommonsorglicensesby30)

Appetite 84 (2015) 1ndash6

Contents lists available at ScienceDirect

Appetite

journal homepage wwwelseviercom locate appet

familiarizing toddlers and preschoolers with the sensory qualitiesof food Popular programs include Ellarsquos Explorers (Ellarsquos Kitchen2011) and Taste for Life (Organix 2010) developed from the methodsproposed in a range of Mange Tout books (Thomas 2007) Whilethese programs are based on the personal experiences of the authorsrather than academic theory or evidence the activities have provenvery popular and several thousand nurseries have adopted them intotheir daily curriculum There are therefore good grounds to thinkthat non-taste activities involving fruits and vegetables would bewelcomed by parents and caregivers should the evidence confirmthat these activities do indeed support healthy eating

To date little research has investigated the role played by theindividual non-taste senses in childrenrsquos food acceptance (seeDazeley Houston-Price amp Hill 2012 for a review) There is encour-aging evidence to suggest that visual and olfactory exposure canpromote consumption (Bronstein amp Crockett 1976 Fedoroff Polivyamp Herman 1997 Heath Houston-Price amp Kennedy 2014 HennessySmotherman amp Levine 1977 Houston-Price Butler amp Shiba 2009)although olfactory exposure is yet to be explored as a means of in-creasing fruit and vegetable intake in young children The effectsof familiarization with the texture andor sound qualities of fruitand vegetables have not as yet been investigated

A small body of work has looked into the effects of holistic sense-based interventions with school-aged children aged between 6 and11 years For example Le goucirct de lrsquoenfant classes educate childrento use their senses when eating and drinking (Puisais amp Pierre 1987)a lsquotastersquo lesson might teach children to discriminate between drinkswith or without added sugar for example The popularity of theprogram has led to its translation from French to Swedish (Hagmanamp Algotson 2000) and there is some evidence of effectiveness Inone study parents reported that childrenrsquos eating behavior was lessneophobic after they participated in the Le goucirct de lrsquoenfant classesand that they had tried a wider variety of foods than a control group(Mustonen amp Tuorila 2010) However a study that followed chil-dren up 10 months later found that childrenrsquos neophobia levels andwillingness to taste novel foods had returned to baseline levels bythis point (Reverdy Chesnel Schlich Koumlster amp Lange 2008)

It is difficult however to build hypotheses about the likely successof sense-based activities with nursery-school-aged children on thebasis of this work While Mustonen Rantanen and Tuorila (2009)reported that it was the children at the younger end of the 6- to11-year-old age range targeted by the Le goucirct de lrsquoenfant classes whobenefited most from participation there has to date been no in-vestigation into the effectiveness of sensory activities designed forpre-schoolers Yet the first few years of life are critical for devel-oping life-long food preferences (Harris 2008) and sensoryinteractions with foods at an early age may have a profound andlasting impact on eating habits

Second the Le goucirct de lrsquoenfant program does not specifically targethealthy foods and so may or may not increase consumption of thefood groups promoted in healthy eating guidelines such as fruitsand vegetables Additionally the lack of appropriate control groupsin many studies of the Le goucirct de lrsquoenfant program complicates in-terpretation of their findings Given the known impact of socio-economic and demographic factors on childrenrsquos diets (Whichelowamp Prevost 1996) it is vital that studies recruit experimental andcontrol participants from the same pool

In sum there is a sparse but encouraging body of evidence tosupport exposure to the sensory qualities of fruits and vegetablesas a means of encouraging consumption We hypothesized that en-couraging children to engage in sense-based playtime activities withunfamiliar fruit and vegetables would increase their willingness toconsume the foods at a later mealtime setting Exposure activitieswere similar to those already adopted by nurseries and includedlooking at and drawing pictures of a food feeling its external andinternal texture smelling it listening to the sound it makes

when squashed or snapped and hearing its name Given howchallenging caregivers find it to encourage young children to tastefoods (Carruth et al 2004) taste exposure was not included in thisstudy The aim was not to tease apart the individual effects of fa-miliarization with a food in the different sense modalities but totest the effects of a holistic sense-based approach to food famil-iarization with children under the age of 3 years as it might beapplied in a day-care or home setting

Method

Participants

Twelve nursery classes were selected for the intervention twofrom each of six privately-owned day care nurseries in BerkshireSix classes included 12- to 24-month-old children while six classesincluded 24- to 36-month-old children Three classes from each agegroup were randomly assigned to form the experimental group theremaining three classes comprised the control group A total of 121children were recruited of whom 92 completed the test session55 in the experimental condition (mean age = 2 0 range = 1 1ndash211) and 37 in the control group (mean age = 2 0 range = 1 0ndash29) The remaining 29 children were absent on the day of testinglargely due to family holidays Informed consent was provided bythe parents of all participants

Materials

Exposure activitiesEach nursery was provided with four activity sheets each ex-

plaining three games specific to one sense sight smell touch orsound (see Appendix for a list of activities) Activities were devisedin collaboration with a childcare professional to ensure that theywere suitable for toddlers and complemented the national curric-ulum Additional resources to support the delivery of the activitiesincluded a story and nursery rhyme about each food and a lami-nated A4 colored photo of each food

Target foodsParticipants were exposed to one of two sets of four foods two

fruits and two vegetables Set A foods were sweet potato greenpepper rhubarb and dried figs Set B foods were butternut squashbroad beans dried prunes and pomegranates These foods were se-lected as likely to be unfamiliar to most children based on the reportsof the nursery staff about the foods served at lunchtime and on theresponses of parents of similarly-aged children to a Fruit and Veg-etable Preference and Familiarity Questionnaire in a previous study(Heath 2013) The two classes at each nursery were randomly al-located to one of the two sets of foods for the exposure phase tocontrol for the possible greater familiarity of one set of foods at oneschool Foods were chosen for their diverse sensory qualities (shapecolor smell and texture) and the two sets of foods were approxi-mately matched for calorie content and colorfulness Foods weredelivered to each nursery once a week by the schoolrsquos regularsupplier

For the exposure phase foods were prepared on site (raw orcooked as required by the activity) by qualified kitchen staff whowere instructed to dice raw foods into 2 cm3 portions and to preparecooked foods as if they were to be served as part of a nursery mealFor the test phase foods were prepared in the same way and wereoffered raw (green pepper dried figs dried prunes pomegran-ates) or cooked (sweet potato rhubarb butternut squash broadbeans) as appropriate on the plates usually used at mealtimes Avery small amount of sugar was added when rhubarb was pre-pared (1 tsp per lb of fruit) in order to make it palatable

2 P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

RecordingA video camera was used to record all test sessions for later

coding

Procedure

Exposure phaseThe exposure phase lasted 4 weeks with each week following

the same schedule On 4 days of the week (either Monday to Thurs-day or Tuesday to Friday) children in the experimental groups spent5ndash10 minutes as a group completing three food-related activitiesfocused on one of the senses (Sight Sound Smell or Touch seeAppendix) The day on which each of the senses was targeted wasrandomly determined for each nursery but was the same each weekAs children did not all attend nursery every day the specific ac-tivities each child experienced depended on the days they attendednursery Half of the experimental groups completed the activitieswith Set A foods the remaining children completed the activitieswith Set B foods

Activities were delivered by key workers at each nursery Ses-sions typically involved six children and three staff one of whomled the session while the others ensured that children were engagedLarger nurseries ran two or more groups simultaneously Activi-ties took place in site-appropriate locations within the nurseryincluding inside the nursery around a table in the story-time areain the soft-play area or in the garden as appropriate for the ac-tivity Activities were deliberately never conducted in a mealtimecontext

To ensure consistency of the delivery of activities staff weretrained the day before the exposure phase started and were as-sisted by a researcher on the first day of the intervention Staff wereinstructed to frame each dayrsquos activities around the specific sensetargeted that day (eg ldquoToday wersquore going to have fun using ourearsrdquo) and to ensure that all discussion about the foods and theirsensory properties was positive regardless of any personal viewsthey might hold about a food During activities that gave childrendirect access to a food tasting was discouraged although staff wereinstructed to avoid saying anything negative if a child ate a foodNursery workers were encouraged to engage the children with eachof the activities for up to 10 minutes but to end the activities if thechildren lost interest While the youngest children were often unableto participate fully in the activities nursery staff ensured they wereinvolved as much as possible (eg they were provided with crayonsand paper even if unable to draw a coherent representation of thefood)

Children in the control group did not take part in any exposureactivities

Test phaseDuring the week immediately following the final week of the

exposure phase children in the experimental and control groupswere tested on their willingness to consume the Set A and Set Bfoods using a two-alternative forced choice procedure Children weretested individually at the dining table where they usually had theirmeals testing took place throughout the day but not during or im-mediately before or after lunch

Participants were presented with four plates of food one at atime Each plate contained two pieces of two different foods onefrom Set A and one from Set B cut to a suitable size for partici-pants to eat Foods were paired as follows butternut squash andsweet potato green beans and green peppers prunes and figsrhubarb and pomegranate For children in the experimental groupsthe target food on each trial depended on the set of foods to whichthey had been exposed The two foods on each plate were namedas they were presented to the child who was asked which food heor she would like to taste If they ate a piece of food they were asked

if they would like any more before being asked if they would liketo try the other food The plate was removed when all the food waseaten or after 2 minutes if the food was refused This procedurewas repeated for the remaining three plates until all eight foods hadbeen offered

The order of presentation of the four plates was counterbal-anced between participants using a Latin Square The location ofthe target food on the left or right side of the plate was counter-balanced within participants

The testing procedure was led by a member of the nursery staffto emulate the natural mealtime environment and to encourage chil-dren to feel comfortable about eating the foods offered The staffmember had not been involved in the exposure activities and wasblind to the set of foods to which children in the experimental groupshad been exposed Staff were again trained in advance of the testphase to ensure consistency of the procedure If a child refused toeat anything staff were instructed to offer gentle encouragementas they would normally at mealtimes but to not put pressure onthe child to try either food The researcher prepared the foods forthe test phase and observed and recorded all sessions Prior to testingparticipants in the control condition met the researcher during aplay session in their classroom to ensure that they were not in-timidated by his presence during testing

CodingThe researcher recorded the foods the participants touched and

tasted and in what order lsquoTastingrsquo was defined as touching the foodwith the tongue chewing or swallowing was not required Videorecordings of test sessions were reviewed to double check that nobehaviors had been missed during the online coding

Results

Mean numbers of Set A and Set B foods touched and tasted bythe children in the experimental and control groups are presentedin Table 1 Pairwise comparisons were employed to establish whetherthe children in each group were more willing to touch or taste SetA or Set B foods Parametric analyses (related t-tests) were em-ployed for tests of childrenrsquos behavior toward fruit and vegetablescombined but non-parametric (Wilcoxon) tests were used for testsof childrenrsquos behavior toward fruits and vegetables separately dueto the small numbers of foods included in these analyses

Children in the control group showed no preference for Set Aor Set B foods in the numbers of foods they touched or tasted (allps gt 05) In contrast the children who were exposed to Set A foodstouched significantly more Set A than Set B vegetables whilechildren who were exposed to Set B foods touched and tasted sig-nificantly more of the Set B vegetables The most meaningful resultsare those from the two exposure groups combined which controlfor any differences in the natural appeal of the two sets of foodsThis overall analysis confirmed that children in the experimentalgroups touched significantly more of the exposed than non-exposed foods t(53) = 205 p = 046 indicating that the exposureactivities had increased childrenrsquos willingness to handle the tar-geted foods This effect was driven by childrenrsquos greater willingnessto touch exposed than non-exposed vegetables (Z = 287 p = 004)no significant effect was seen for fruit (Z = 022 p = 83) A similarbut weaker pattern was seen in childrenrsquos tasting behavior Whilechildren did not taste significantly more of the exposed foods overallt(53) = 165 p = 11 they tasted more of the exposed vegetables thanof the non-exposed vegetables (Z = 224 p = 025) again no effectwas found for fruit (p = 1)

Childrenrsquos greater willingness to touch and taste exposed foodswas also reflected in the order in which they interacted with thefoods on each plate Table 2 presents the mean numbers of Set Aand Set B foods that children touched or tasted before they

3P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

approached the alternative food on each plate Children who hadbeen exposed to Set A foods systematically touched these beforethey touched the non-exposed foods t(16) = 250 p = 023 they alsotasted the Set A fruits before they tasted the non-exposed fruits(Z = 212 p = 034) In contrast children who had been exposed toSet B foods tasted the Set B foods first t(13) = 219 p = 047 The ex-posure effect for this group was most strongly seen in childrenrsquosbehavior toward vegetables the exposed vegetables were bothtouched (Z = 284 p = 005) and tasted (Z = 258 p = 010) before thenon-exposed vegetables

It is important to note that children in the control group alsotouched (Z = 329 p = 001) and tasted (Z = 308 p = 002) the Set Bfoods before the Set A foods despite their lack of exposure to thesesuggesting that foods in Set B were more naturally appealing Thecrucial comparisons control for these differences between the twosets of foods by comparing the behavior of the two experimentalgroups combined toward exposed and non-exposed foods As canbe seen in Table 2 children as a group touched the foods to whichthey had been exposed before those to which they had not beenexposed t(36) = 213 p = 040 this pattern was driven by their be-havior toward vegetables (Z = 204 p = 042) rather than fruits(Z = 036 p = 72) Importantly children also tasted the exposed foodsbefore the non-exposed foods t(23) = 229 p = 032 and in this casechildrenrsquos behavior was driven by their reactions toward fruits(Z = 256 p = 011) rather than vegetables (Z = 142 p = 16)

Discussion

This study provides the first investigation into the effects of in-troducing new fruits and vegetables in playtime activities on toddlersrsquosubsequent acceptance of the foods at a mealtime setting Resultssuggest that familiarizing children with the non-taste sensory quali-ties of foods increases childrenrsquos willingness to touch and taste themwhen they are later offered these to eat Specifically when the totalnumbers of exposed and non-exposed foods that children werewilling to touch or taste were compared the intervention was seento impact strongly on childrenrsquos behavior toward vegetables Tod-dlers both touched and tasted more of the exposed than non-exposed vegetables they also touched more of the exposed foodsthan non-exposed foods overall

The order in which children approached the foods on each platealso demonstrated their greater confidence in engaging with the pre-viously exposed foods children touched and tasted the exposedfoods before they touched and tasted the non-exposed foods Chil-drenrsquos greater willingness to engage with the exposed foods wasevident toward both fruits and vegetables on this measure chil-dren touched the vegetable to which they had been exposed beforethe non-exposed vegetable on each plate and they tasted theexposed fruit on each plate before the non-exposed fruit

Overall the results demonstrate a very clear impact of the in-tervention on childrenrsquos willingness to try the targeted foods and

Table 1Numbers of foods in each set touched and tasted by the experimental and control groups

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched 4 267 242 197 233 208 222 248 217SD 124 135 145 147 180 173 137 141Mean veg touched 2 142 104 93 123 105 108 131 098SD 078 081 083 086 097 089 082 081Mean fruit touched 2 125 138 103 110 103 114 117 119SD 068 071 085 080 093 095 075 080Mean foods tasted 4 146 138 120 142 159 181 144 128SD 128 138 140 148 180 173 138 140Mean veg tasted 2 071 058 057 078 076 086 075 057SD 075 072 078 080 095 089 078 074Mean fruit tasted 2 075 079 061 065 084 095 069 069SD 074 078 076 084 093 091 079 077

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 005)

Table 2Numbers of foods in each set that were touched or tasted before the other food on each plate

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched first 4 159 71 115 135 114 200 146 095SD 080 085 088 104 089 127 093 088Mean veg touched first 2 029 036 018 100 029 132 070 027SD 047 063 039 071 047 058 070 052Mean fruit touched first 2 106 044 107 071 090 090 090 073SD 068 081 062 073 083 083 071 078Mean foods tasted first 4 130 100 050 150 106 206 142 071SD 067 082 065 134 077 118 110 075Mean veg tasted first 2 040 100 021 150 031 206 108 057SD 052 082 043 134 048 118 120 073Mean fruit tasted first 2 090 030 040 100 075 088 095 035SD 032 048 052 067 058 072 051 049

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 01 p lt 005 p lt 001)

4 P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

support the potential for sensory activities to facilitate the intro-duction of healthy foods into childrenrsquos diets

The results of this study suggest several key avenues for furtherresearch First questions remain about the impact of the interven-tion beyond the taste test conducted We can say nothing about thechanges that might have occurred in childrenrsquos eating behavioroutside the nursery environment Future studies might explore theprogramrsquos impact beyond the confines of the nursery setting byasking parents about childrenrsquos eating behavior before and after anintervention at the childrenrsquos nursery Toddlers are likely to havemore influence over what they eat in the home environment thanat nursery and parents may notice if children ask for certain foodsat home or in the supermarket or are more willing to consume theseduring family mealtimes The effects of the intervention might alsobe specific to the foods targeted by the program or they might gen-eralize to other similar foods (Birch Gunder Grimm-Thomas amp Laing1998) or bring about positive changes in childrenrsquos attitudes towardnew foods in general Parents should therefore be asked to reporton childrenrsquos food neophobia andor food fussiness (Pliner 1994Wardle Guthrie Sanderson amp Rapoport 2001) as well as on theirconsumption of specific exposed and non-exposed foods before andafter a nursery intervention is conducted

The longevity of the effects seen in our lsquowillingness to tastersquo testalso remains to be established Previous work has questionedwhether a sense-based program for school-aged children can havea long-term impact on childrenrsquos eating behavior (Reverdy et al2008) In order to justify any recommendation to nurseries to im-plement a program similar to that used in our study we would needto know that the intervention is likely to cause lasting changes inchildrenrsquos willingness to consume the target foods Alternatively ifeffects were found to last for only 6 months or a year for exampleone might recommend repeating the program on a biannual orannual basis to keep childrenrsquos interest in the targeted foods lsquotoppeduprsquo

Other outstanding questions concern the locus of the positiveeffects of the intervention and how these might be optimized Wouldthe sense-based activities we employed have the same impact if theywere conducted in the home environment for example If parentsengaged in food-related activities with their child parentsrsquo own in-terest in and willingness to consume the fruits and vegetablesconcerned might be enhanced along with the childrsquos resulting inpositive changes to the parentrsquos food purchasing and eating behav-ior and hence an even greater impact on childrenrsquos willingness toconsume the targeted foods (Busick Brooks Pernecky Dawson ampPetzoldt 2008 Worobey Ostapkovich Yudin amp Worobey 2010)

It would similarly be beneficial to establish the optimal lengthof the program and whether a program that runs for fewer than 4weeks is just as effective or whether a prolonged but less intenseprogram leads to longer-term changes in food liking Informal feed-back from the staff who participated in the activities suggests thattargeting the sensory activities around a single food each day ratherthan a single sense would result in less food waste and thereforereduce the cost of running the intervention

Finally it would be of both practical and theoretical interest toestablish the contribution of the activitiesrsquo involvement of differ-ent sensory modalities to the positive effects we have reported Whilethe current study has demonstrated the potential for a multisen-sory intervention to impact on childrenrsquos willingness to taste newfoods at mealtimes it cannot speak to the effect of each type of ac-tivity in isolation The participants in our study engaged in differentcombinations of sensory games depending on the days on whichthey attended nursery such that some children took part in gamesinvolving only two of the senses while others experienced activi-ties engaging all four targeted senses However the study was notpowered to investigate whether such differences affected theoutcome for individual children Nor do we know whether within

each sensory modality some activities exerted a greater influenceon childrenrsquos willingness to touch and taste the foods than othersA much larger study in which sub-groups of children were as-signed to different combinations of sensory activities would berequired to tease apart the relative impact of the different sensesin increasing childrenrsquos willingness to taste new foods and to designthe most efficient intervention with the greatest potential to enhancehealthy eating

References

Birch L (1999) Development of food preferences Annual Review of Nutrition 1941ndash62

Birch L Gunder L Grimm-Thomas K amp Laing D (1998) Infantsrsquo consumption ofa new food enhances acceptance of similar foods Appetite 30 283ndash295

Birch L amp Marlin D (1982) I donrsquot like it I never tried it Effects of exposure ontwo-year-old childrenrsquos food preferences Appetite 3 353ndash360

Bronstein P amp Crockett D (1976) Exposure to odor of food determines eatingpreferences of rat pups Behavioral Biology 18 387ndash392

Busick D B Brooks J Pernecky S Dawson R amp Petzoldt J (2008) Parent foodpurchases as a measure of exposure and preschool-aged childrenrsquos willingnessto identify and taste fruit and vegetables Appetite 51 468ndash473

Carruth B Ziegler P Gordon A amp Barr S (2004) Prevalence of picky eaters amonginfants and toddlers and their caregiversrsquo decisions about offering a new foodJournal of the American Dietetic Association 104 S57ndashS64

Cooke L (2007) The importance of exposure for healthy eating in childhoodA review Journal of Human Nutrition amp Dietetics 20 294ndash301

Dazeley P Houston-Price C amp Hill C (2012) Should healthy eating programmesincorporate interaction with foods in different sensory modalities A review ofthe evidence British Journal of Nutrition 108 769ndash777

Desor J Maller O amp Andrews K (1975) Ingestive responses of human newbornsto salty sour and bitter stimuli Journal of Comparative amp Physiological Psychology89 966ndash970

Desor J Maller O amp Turner R (1973) Taste in acceptance of sugars by humaninfants Journal of Comparative amp Physiological Psychology 3 496ndash501

Ellarsquos Kitchen (2011) Ellarsquos Explorers lthttpwwwellaskitchencoukare-you-a-nurserygt Last accessed 140312

Fedoroff I Polivy J amp Herman P (1997) The effect of pre-exposure to food cueson the eating behavior of restrained and unrestrained eaters Appetite 28 33ndash47

Gerrish C amp Mennella J (2001) Flavor variety enhances food acceptance informula-fed infants American Journal of Clinical Nutrition 73 1080ndash1085

Hagman U amp Algotson S (2000) Mat foumlr alla sinnen-sensorisk traumlning enligt SAPEREmetoden Stockholm Blomberg amp Jansson

Harris G (2008) Development of taste and food preferences in children CurrentOpinion in Clinical Nutrition amp Metabolic Care 11 315ndash319

Heath P (2013) Improving childrenrsquos responses to fruit and vegetables Picture-bookexposure and the impact of food familiarity and liking (PhD thesis) Universityof Reading

Heath P Houston-Price C amp Kennedy O B (2014) Letrsquos look at leeks Picture booksincrease toddlersrsquo willingness to look at taste and consume unfamiliar vegetablesFrontiers in Psychology 5(191) 1ndash11 doi103389fpsyg201400191

Hennessy M Smotherman W amp Levine S (1977) Early olfactory enrichmentenhances later consumption of novel substances Physiology amp Behaviour 19481ndash483

Houston-Price C Butler L amp Shiba P (2009) Visual exposure impacts on toddlersrsquowillingness to taste fruits and vegetables Appetite 53 450ndash453

Kalat J amp Rozin P (1973) ldquoLearned safetyrdquo as a mechanism in long-delay taste-aversion learning in rats Journal of Comparative amp Physiological Psychology 83198ndash207

Mustonen S Rantanen R amp Tuorila H (2009) Effect of sensory education on schoolchildrenrsquos food perception A 2-year follow-up study Food Quality amp Preference20 230ndash240

Mustonen S amp Tuorila H (2010) Sensory education decreases food neophobia scoreand encourages trying unfamiliar foods in 8ndash12-year-old children Food Qualityamp Preference 21 353ndash360

Organix (2010) Taste for Life lthttpwwworganixcomsupport-advicetaste-lifegtLast accessed 201010

Pliner P (1994) Development of measures of food neophobia in children Appetite23 147ndash163

Puisais J amp Pierre C (1987) Le goucirct de lrsquoenfant Paris FlammarionReverdy C Chesnel F Schlich P Koumlster E amp Lange C (2008) Effect of sensory

education on willingness to taste novel food in children Appetite 51 156ndash165Rozin P (1976) The selection of food by rats humans and other animals In J

Rosenblatt R Hinde C Beer amp E Shaw (Eds) Advances in the study of behaviour(Vol 6 pp 21ndash76) LondonNew York Academic Books

Thomas L (2007) Mange tout Teaching your children to love fruit and vegetableswithout tears London Penguin

Wardle J amp Cooke L (2008) Genetic and environmental determinants of childrenrsquosfood preferences British Journal of Nutrition 99 S15ndashS21

Wardle J Cooke L Gibson E L Sapochnik M Sheiham A amp Lawson M (2003a)Increasing childrenrsquos acceptance of vegetables A randomized trial of parent-ledexposure Appetite 40 155ndash162

5P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

RecordingA video camera was used to record all test sessions for later

coding

Procedure

Exposure phaseThe exposure phase lasted 4 weeks with each week following

the same schedule On 4 days of the week (either Monday to Thurs-day or Tuesday to Friday) children in the experimental groups spent5ndash10 minutes as a group completing three food-related activitiesfocused on one of the senses (Sight Sound Smell or Touch seeAppendix) The day on which each of the senses was targeted wasrandomly determined for each nursery but was the same each weekAs children did not all attend nursery every day the specific ac-tivities each child experienced depended on the days they attendednursery Half of the experimental groups completed the activitieswith Set A foods the remaining children completed the activitieswith Set B foods

Activities were delivered by key workers at each nursery Ses-sions typically involved six children and three staff one of whomled the session while the others ensured that children were engagedLarger nurseries ran two or more groups simultaneously Activi-ties took place in site-appropriate locations within the nurseryincluding inside the nursery around a table in the story-time areain the soft-play area or in the garden as appropriate for the ac-tivity Activities were deliberately never conducted in a mealtimecontext

To ensure consistency of the delivery of activities staff weretrained the day before the exposure phase started and were as-sisted by a researcher on the first day of the intervention Staff wereinstructed to frame each dayrsquos activities around the specific sensetargeted that day (eg ldquoToday wersquore going to have fun using ourearsrdquo) and to ensure that all discussion about the foods and theirsensory properties was positive regardless of any personal viewsthey might hold about a food During activities that gave childrendirect access to a food tasting was discouraged although staff wereinstructed to avoid saying anything negative if a child ate a foodNursery workers were encouraged to engage the children with eachof the activities for up to 10 minutes but to end the activities if thechildren lost interest While the youngest children were often unableto participate fully in the activities nursery staff ensured they wereinvolved as much as possible (eg they were provided with crayonsand paper even if unable to draw a coherent representation of thefood)

Children in the control group did not take part in any exposureactivities

Test phaseDuring the week immediately following the final week of the

exposure phase children in the experimental and control groupswere tested on their willingness to consume the Set A and Set Bfoods using a two-alternative forced choice procedure Children weretested individually at the dining table where they usually had theirmeals testing took place throughout the day but not during or im-mediately before or after lunch

Participants were presented with four plates of food one at atime Each plate contained two pieces of two different foods onefrom Set A and one from Set B cut to a suitable size for partici-pants to eat Foods were paired as follows butternut squash andsweet potato green beans and green peppers prunes and figsrhubarb and pomegranate For children in the experimental groupsthe target food on each trial depended on the set of foods to whichthey had been exposed The two foods on each plate were namedas they were presented to the child who was asked which food heor she would like to taste If they ate a piece of food they were asked

if they would like any more before being asked if they would liketo try the other food The plate was removed when all the food waseaten or after 2 minutes if the food was refused This procedurewas repeated for the remaining three plates until all eight foods hadbeen offered

The order of presentation of the four plates was counterbal-anced between participants using a Latin Square The location ofthe target food on the left or right side of the plate was counter-balanced within participants

The testing procedure was led by a member of the nursery staffto emulate the natural mealtime environment and to encourage chil-dren to feel comfortable about eating the foods offered The staffmember had not been involved in the exposure activities and wasblind to the set of foods to which children in the experimental groupshad been exposed Staff were again trained in advance of the testphase to ensure consistency of the procedure If a child refused toeat anything staff were instructed to offer gentle encouragementas they would normally at mealtimes but to not put pressure onthe child to try either food The researcher prepared the foods forthe test phase and observed and recorded all sessions Prior to testingparticipants in the control condition met the researcher during aplay session in their classroom to ensure that they were not in-timidated by his presence during testing

CodingThe researcher recorded the foods the participants touched and

tasted and in what order lsquoTastingrsquo was defined as touching the foodwith the tongue chewing or swallowing was not required Videorecordings of test sessions were reviewed to double check that nobehaviors had been missed during the online coding

Results

Mean numbers of Set A and Set B foods touched and tasted bythe children in the experimental and control groups are presentedin Table 1 Pairwise comparisons were employed to establish whetherthe children in each group were more willing to touch or taste SetA or Set B foods Parametric analyses (related t-tests) were em-ployed for tests of childrenrsquos behavior toward fruit and vegetablescombined but non-parametric (Wilcoxon) tests were used for testsof childrenrsquos behavior toward fruits and vegetables separately dueto the small numbers of foods included in these analyses

Children in the control group showed no preference for Set Aor Set B foods in the numbers of foods they touched or tasted (allps gt 05) In contrast the children who were exposed to Set A foodstouched significantly more Set A than Set B vegetables whilechildren who were exposed to Set B foods touched and tasted sig-nificantly more of the Set B vegetables The most meaningful resultsare those from the two exposure groups combined which controlfor any differences in the natural appeal of the two sets of foodsThis overall analysis confirmed that children in the experimentalgroups touched significantly more of the exposed than non-exposed foods t(53) = 205 p = 046 indicating that the exposureactivities had increased childrenrsquos willingness to handle the tar-geted foods This effect was driven by childrenrsquos greater willingnessto touch exposed than non-exposed vegetables (Z = 287 p = 004)no significant effect was seen for fruit (Z = 022 p = 83) A similarbut weaker pattern was seen in childrenrsquos tasting behavior Whilechildren did not taste significantly more of the exposed foods overallt(53) = 165 p = 11 they tasted more of the exposed vegetables thanof the non-exposed vegetables (Z = 224 p = 025) again no effectwas found for fruit (p = 1)

Childrenrsquos greater willingness to touch and taste exposed foodswas also reflected in the order in which they interacted with thefoods on each plate Table 2 presents the mean numbers of Set Aand Set B foods that children touched or tasted before they

3P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

approached the alternative food on each plate Children who hadbeen exposed to Set A foods systematically touched these beforethey touched the non-exposed foods t(16) = 250 p = 023 they alsotasted the Set A fruits before they tasted the non-exposed fruits(Z = 212 p = 034) In contrast children who had been exposed toSet B foods tasted the Set B foods first t(13) = 219 p = 047 The ex-posure effect for this group was most strongly seen in childrenrsquosbehavior toward vegetables the exposed vegetables were bothtouched (Z = 284 p = 005) and tasted (Z = 258 p = 010) before thenon-exposed vegetables

It is important to note that children in the control group alsotouched (Z = 329 p = 001) and tasted (Z = 308 p = 002) the Set Bfoods before the Set A foods despite their lack of exposure to thesesuggesting that foods in Set B were more naturally appealing Thecrucial comparisons control for these differences between the twosets of foods by comparing the behavior of the two experimentalgroups combined toward exposed and non-exposed foods As canbe seen in Table 2 children as a group touched the foods to whichthey had been exposed before those to which they had not beenexposed t(36) = 213 p = 040 this pattern was driven by their be-havior toward vegetables (Z = 204 p = 042) rather than fruits(Z = 036 p = 72) Importantly children also tasted the exposed foodsbefore the non-exposed foods t(23) = 229 p = 032 and in this casechildrenrsquos behavior was driven by their reactions toward fruits(Z = 256 p = 011) rather than vegetables (Z = 142 p = 16)

Discussion

This study provides the first investigation into the effects of in-troducing new fruits and vegetables in playtime activities on toddlersrsquosubsequent acceptance of the foods at a mealtime setting Resultssuggest that familiarizing children with the non-taste sensory quali-ties of foods increases childrenrsquos willingness to touch and taste themwhen they are later offered these to eat Specifically when the totalnumbers of exposed and non-exposed foods that children werewilling to touch or taste were compared the intervention was seento impact strongly on childrenrsquos behavior toward vegetables Tod-dlers both touched and tasted more of the exposed than non-exposed vegetables they also touched more of the exposed foodsthan non-exposed foods overall

The order in which children approached the foods on each platealso demonstrated their greater confidence in engaging with the pre-viously exposed foods children touched and tasted the exposedfoods before they touched and tasted the non-exposed foods Chil-drenrsquos greater willingness to engage with the exposed foods wasevident toward both fruits and vegetables on this measure chil-dren touched the vegetable to which they had been exposed beforethe non-exposed vegetable on each plate and they tasted theexposed fruit on each plate before the non-exposed fruit

Overall the results demonstrate a very clear impact of the in-tervention on childrenrsquos willingness to try the targeted foods and

Table 1Numbers of foods in each set touched and tasted by the experimental and control groups

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched 4 267 242 197 233 208 222 248 217SD 124 135 145 147 180 173 137 141Mean veg touched 2 142 104 93 123 105 108 131 098SD 078 081 083 086 097 089 082 081Mean fruit touched 2 125 138 103 110 103 114 117 119SD 068 071 085 080 093 095 075 080Mean foods tasted 4 146 138 120 142 159 181 144 128SD 128 138 140 148 180 173 138 140Mean veg tasted 2 071 058 057 078 076 086 075 057SD 075 072 078 080 095 089 078 074Mean fruit tasted 2 075 079 061 065 084 095 069 069SD 074 078 076 084 093 091 079 077

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 005)

Table 2Numbers of foods in each set that were touched or tasted before the other food on each plate

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched first 4 159 71 115 135 114 200 146 095SD 080 085 088 104 089 127 093 088Mean veg touched first 2 029 036 018 100 029 132 070 027SD 047 063 039 071 047 058 070 052Mean fruit touched first 2 106 044 107 071 090 090 090 073SD 068 081 062 073 083 083 071 078Mean foods tasted first 4 130 100 050 150 106 206 142 071SD 067 082 065 134 077 118 110 075Mean veg tasted first 2 040 100 021 150 031 206 108 057SD 052 082 043 134 048 118 120 073Mean fruit tasted first 2 090 030 040 100 075 088 095 035SD 032 048 052 067 058 072 051 049

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 01 p lt 005 p lt 001)

4 P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

support the potential for sensory activities to facilitate the intro-duction of healthy foods into childrenrsquos diets

The results of this study suggest several key avenues for furtherresearch First questions remain about the impact of the interven-tion beyond the taste test conducted We can say nothing about thechanges that might have occurred in childrenrsquos eating behavioroutside the nursery environment Future studies might explore theprogramrsquos impact beyond the confines of the nursery setting byasking parents about childrenrsquos eating behavior before and after anintervention at the childrenrsquos nursery Toddlers are likely to havemore influence over what they eat in the home environment thanat nursery and parents may notice if children ask for certain foodsat home or in the supermarket or are more willing to consume theseduring family mealtimes The effects of the intervention might alsobe specific to the foods targeted by the program or they might gen-eralize to other similar foods (Birch Gunder Grimm-Thomas amp Laing1998) or bring about positive changes in childrenrsquos attitudes towardnew foods in general Parents should therefore be asked to reporton childrenrsquos food neophobia andor food fussiness (Pliner 1994Wardle Guthrie Sanderson amp Rapoport 2001) as well as on theirconsumption of specific exposed and non-exposed foods before andafter a nursery intervention is conducted

The longevity of the effects seen in our lsquowillingness to tastersquo testalso remains to be established Previous work has questionedwhether a sense-based program for school-aged children can havea long-term impact on childrenrsquos eating behavior (Reverdy et al2008) In order to justify any recommendation to nurseries to im-plement a program similar to that used in our study we would needto know that the intervention is likely to cause lasting changes inchildrenrsquos willingness to consume the target foods Alternatively ifeffects were found to last for only 6 months or a year for exampleone might recommend repeating the program on a biannual orannual basis to keep childrenrsquos interest in the targeted foods lsquotoppeduprsquo

Other outstanding questions concern the locus of the positiveeffects of the intervention and how these might be optimized Wouldthe sense-based activities we employed have the same impact if theywere conducted in the home environment for example If parentsengaged in food-related activities with their child parentsrsquo own in-terest in and willingness to consume the fruits and vegetablesconcerned might be enhanced along with the childrsquos resulting inpositive changes to the parentrsquos food purchasing and eating behav-ior and hence an even greater impact on childrenrsquos willingness toconsume the targeted foods (Busick Brooks Pernecky Dawson ampPetzoldt 2008 Worobey Ostapkovich Yudin amp Worobey 2010)

It would similarly be beneficial to establish the optimal lengthof the program and whether a program that runs for fewer than 4weeks is just as effective or whether a prolonged but less intenseprogram leads to longer-term changes in food liking Informal feed-back from the staff who participated in the activities suggests thattargeting the sensory activities around a single food each day ratherthan a single sense would result in less food waste and thereforereduce the cost of running the intervention

Finally it would be of both practical and theoretical interest toestablish the contribution of the activitiesrsquo involvement of differ-ent sensory modalities to the positive effects we have reported Whilethe current study has demonstrated the potential for a multisen-sory intervention to impact on childrenrsquos willingness to taste newfoods at mealtimes it cannot speak to the effect of each type of ac-tivity in isolation The participants in our study engaged in differentcombinations of sensory games depending on the days on whichthey attended nursery such that some children took part in gamesinvolving only two of the senses while others experienced activi-ties engaging all four targeted senses However the study was notpowered to investigate whether such differences affected theoutcome for individual children Nor do we know whether within

each sensory modality some activities exerted a greater influenceon childrenrsquos willingness to touch and taste the foods than othersA much larger study in which sub-groups of children were as-signed to different combinations of sensory activities would berequired to tease apart the relative impact of the different sensesin increasing childrenrsquos willingness to taste new foods and to designthe most efficient intervention with the greatest potential to enhancehealthy eating

References

Birch L (1999) Development of food preferences Annual Review of Nutrition 1941ndash62

Birch L Gunder L Grimm-Thomas K amp Laing D (1998) Infantsrsquo consumption ofa new food enhances acceptance of similar foods Appetite 30 283ndash295

Birch L amp Marlin D (1982) I donrsquot like it I never tried it Effects of exposure ontwo-year-old childrenrsquos food preferences Appetite 3 353ndash360

Bronstein P amp Crockett D (1976) Exposure to odor of food determines eatingpreferences of rat pups Behavioral Biology 18 387ndash392

Busick D B Brooks J Pernecky S Dawson R amp Petzoldt J (2008) Parent foodpurchases as a measure of exposure and preschool-aged childrenrsquos willingnessto identify and taste fruit and vegetables Appetite 51 468ndash473

Carruth B Ziegler P Gordon A amp Barr S (2004) Prevalence of picky eaters amonginfants and toddlers and their caregiversrsquo decisions about offering a new foodJournal of the American Dietetic Association 104 S57ndashS64

Cooke L (2007) The importance of exposure for healthy eating in childhoodA review Journal of Human Nutrition amp Dietetics 20 294ndash301

Dazeley P Houston-Price C amp Hill C (2012) Should healthy eating programmesincorporate interaction with foods in different sensory modalities A review ofthe evidence British Journal of Nutrition 108 769ndash777

Desor J Maller O amp Andrews K (1975) Ingestive responses of human newbornsto salty sour and bitter stimuli Journal of Comparative amp Physiological Psychology89 966ndash970

Desor J Maller O amp Turner R (1973) Taste in acceptance of sugars by humaninfants Journal of Comparative amp Physiological Psychology 3 496ndash501

Ellarsquos Kitchen (2011) Ellarsquos Explorers lthttpwwwellaskitchencoukare-you-a-nurserygt Last accessed 140312

Fedoroff I Polivy J amp Herman P (1997) The effect of pre-exposure to food cueson the eating behavior of restrained and unrestrained eaters Appetite 28 33ndash47

Gerrish C amp Mennella J (2001) Flavor variety enhances food acceptance informula-fed infants American Journal of Clinical Nutrition 73 1080ndash1085

Hagman U amp Algotson S (2000) Mat foumlr alla sinnen-sensorisk traumlning enligt SAPEREmetoden Stockholm Blomberg amp Jansson

Harris G (2008) Development of taste and food preferences in children CurrentOpinion in Clinical Nutrition amp Metabolic Care 11 315ndash319

Heath P (2013) Improving childrenrsquos responses to fruit and vegetables Picture-bookexposure and the impact of food familiarity and liking (PhD thesis) Universityof Reading

Heath P Houston-Price C amp Kennedy O B (2014) Letrsquos look at leeks Picture booksincrease toddlersrsquo willingness to look at taste and consume unfamiliar vegetablesFrontiers in Psychology 5(191) 1ndash11 doi103389fpsyg201400191

Hennessy M Smotherman W amp Levine S (1977) Early olfactory enrichmentenhances later consumption of novel substances Physiology amp Behaviour 19481ndash483

Houston-Price C Butler L amp Shiba P (2009) Visual exposure impacts on toddlersrsquowillingness to taste fruits and vegetables Appetite 53 450ndash453

Kalat J amp Rozin P (1973) ldquoLearned safetyrdquo as a mechanism in long-delay taste-aversion learning in rats Journal of Comparative amp Physiological Psychology 83198ndash207

Mustonen S Rantanen R amp Tuorila H (2009) Effect of sensory education on schoolchildrenrsquos food perception A 2-year follow-up study Food Quality amp Preference20 230ndash240

Mustonen S amp Tuorila H (2010) Sensory education decreases food neophobia scoreand encourages trying unfamiliar foods in 8ndash12-year-old children Food Qualityamp Preference 21 353ndash360

Organix (2010) Taste for Life lthttpwwworganixcomsupport-advicetaste-lifegtLast accessed 201010

Pliner P (1994) Development of measures of food neophobia in children Appetite23 147ndash163

Puisais J amp Pierre C (1987) Le goucirct de lrsquoenfant Paris FlammarionReverdy C Chesnel F Schlich P Koumlster E amp Lange C (2008) Effect of sensory

education on willingness to taste novel food in children Appetite 51 156ndash165Rozin P (1976) The selection of food by rats humans and other animals In J

Rosenblatt R Hinde C Beer amp E Shaw (Eds) Advances in the study of behaviour(Vol 6 pp 21ndash76) LondonNew York Academic Books

Thomas L (2007) Mange tout Teaching your children to love fruit and vegetableswithout tears London Penguin

Wardle J amp Cooke L (2008) Genetic and environmental determinants of childrenrsquosfood preferences British Journal of Nutrition 99 S15ndashS21

Wardle J Cooke L Gibson E L Sapochnik M Sheiham A amp Lawson M (2003a)Increasing childrenrsquos acceptance of vegetables A randomized trial of parent-ledexposure Appetite 40 155ndash162

5P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

approached the alternative food on each plate Children who hadbeen exposed to Set A foods systematically touched these beforethey touched the non-exposed foods t(16) = 250 p = 023 they alsotasted the Set A fruits before they tasted the non-exposed fruits(Z = 212 p = 034) In contrast children who had been exposed toSet B foods tasted the Set B foods first t(13) = 219 p = 047 The ex-posure effect for this group was most strongly seen in childrenrsquosbehavior toward vegetables the exposed vegetables were bothtouched (Z = 284 p = 005) and tasted (Z = 258 p = 010) before thenon-exposed vegetables

It is important to note that children in the control group alsotouched (Z = 329 p = 001) and tasted (Z = 308 p = 002) the Set Bfoods before the Set A foods despite their lack of exposure to thesesuggesting that foods in Set B were more naturally appealing Thecrucial comparisons control for these differences between the twosets of foods by comparing the behavior of the two experimentalgroups combined toward exposed and non-exposed foods As canbe seen in Table 2 children as a group touched the foods to whichthey had been exposed before those to which they had not beenexposed t(36) = 213 p = 040 this pattern was driven by their be-havior toward vegetables (Z = 204 p = 042) rather than fruits(Z = 036 p = 72) Importantly children also tasted the exposed foodsbefore the non-exposed foods t(23) = 229 p = 032 and in this casechildrenrsquos behavior was driven by their reactions toward fruits(Z = 256 p = 011) rather than vegetables (Z = 142 p = 16)

Discussion

This study provides the first investigation into the effects of in-troducing new fruits and vegetables in playtime activities on toddlersrsquosubsequent acceptance of the foods at a mealtime setting Resultssuggest that familiarizing children with the non-taste sensory quali-ties of foods increases childrenrsquos willingness to touch and taste themwhen they are later offered these to eat Specifically when the totalnumbers of exposed and non-exposed foods that children werewilling to touch or taste were compared the intervention was seento impact strongly on childrenrsquos behavior toward vegetables Tod-dlers both touched and tasted more of the exposed than non-exposed vegetables they also touched more of the exposed foodsthan non-exposed foods overall

The order in which children approached the foods on each platealso demonstrated their greater confidence in engaging with the pre-viously exposed foods children touched and tasted the exposedfoods before they touched and tasted the non-exposed foods Chil-drenrsquos greater willingness to engage with the exposed foods wasevident toward both fruits and vegetables on this measure chil-dren touched the vegetable to which they had been exposed beforethe non-exposed vegetable on each plate and they tasted theexposed fruit on each plate before the non-exposed fruit

Overall the results demonstrate a very clear impact of the in-tervention on childrenrsquos willingness to try the targeted foods and

Table 1Numbers of foods in each set touched and tasted by the experimental and control groups

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched 4 267 242 197 233 208 222 248 217SD 124 135 145 147 180 173 137 141Mean veg touched 2 142 104 93 123 105 108 131 098SD 078 081 083 086 097 089 082 081Mean fruit touched 2 125 138 103 110 103 114 117 119SD 068 071 085 080 093 095 075 080Mean foods tasted 4 146 138 120 142 159 181 144 128SD 128 138 140 148 180 173 138 140Mean veg tasted 2 071 058 057 078 076 086 075 057SD 075 072 078 080 095 089 078 074Mean fruit tasted 2 075 079 061 065 084 095 069 069SD 074 078 076 084 093 091 079 077

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 005)

Table 2Numbers of foods in each set that were touched or tasted before the other food on each plate

Experimental groupA (N = 24)(Exposed to Set A)

Experimental groupB (N = 31)(Exposed to Set B)

Control group(N = 37)(No exposure)

Experimentalgroups combined(N = 55)

No of foods Set A Set B Set A Set B Set A Set B Exposed Non-exposed

Mean foods touched first 4 159 71 115 135 114 200 146 095SD 080 085 088 104 089 127 093 088Mean veg touched first 2 029 036 018 100 029 132 070 027SD 047 063 039 071 047 058 070 052Mean fruit touched first 2 106 044 107 071 090 090 090 073SD 068 081 062 073 083 083 071 078Mean foods tasted first 4 130 100 050 150 106 206 142 071SD 067 082 065 134 077 118 110 075Mean veg tasted first 2 040 100 021 150 031 206 108 057SD 052 082 043 134 048 118 120 073Mean fruit tasted first 2 090 030 040 100 075 088 095 035SD 032 048 052 067 058 072 051 049

Significant pairwise comparisons between Set A and Set B (or Exposed and Non-exposed) foods are indicated by asterisks (p lt 05 p lt 01 p lt 005 p lt 001)

4 P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

support the potential for sensory activities to facilitate the intro-duction of healthy foods into childrenrsquos diets

The results of this study suggest several key avenues for furtherresearch First questions remain about the impact of the interven-tion beyond the taste test conducted We can say nothing about thechanges that might have occurred in childrenrsquos eating behavioroutside the nursery environment Future studies might explore theprogramrsquos impact beyond the confines of the nursery setting byasking parents about childrenrsquos eating behavior before and after anintervention at the childrenrsquos nursery Toddlers are likely to havemore influence over what they eat in the home environment thanat nursery and parents may notice if children ask for certain foodsat home or in the supermarket or are more willing to consume theseduring family mealtimes The effects of the intervention might alsobe specific to the foods targeted by the program or they might gen-eralize to other similar foods (Birch Gunder Grimm-Thomas amp Laing1998) or bring about positive changes in childrenrsquos attitudes towardnew foods in general Parents should therefore be asked to reporton childrenrsquos food neophobia andor food fussiness (Pliner 1994Wardle Guthrie Sanderson amp Rapoport 2001) as well as on theirconsumption of specific exposed and non-exposed foods before andafter a nursery intervention is conducted

The longevity of the effects seen in our lsquowillingness to tastersquo testalso remains to be established Previous work has questionedwhether a sense-based program for school-aged children can havea long-term impact on childrenrsquos eating behavior (Reverdy et al2008) In order to justify any recommendation to nurseries to im-plement a program similar to that used in our study we would needto know that the intervention is likely to cause lasting changes inchildrenrsquos willingness to consume the target foods Alternatively ifeffects were found to last for only 6 months or a year for exampleone might recommend repeating the program on a biannual orannual basis to keep childrenrsquos interest in the targeted foods lsquotoppeduprsquo

Other outstanding questions concern the locus of the positiveeffects of the intervention and how these might be optimized Wouldthe sense-based activities we employed have the same impact if theywere conducted in the home environment for example If parentsengaged in food-related activities with their child parentsrsquo own in-terest in and willingness to consume the fruits and vegetablesconcerned might be enhanced along with the childrsquos resulting inpositive changes to the parentrsquos food purchasing and eating behav-ior and hence an even greater impact on childrenrsquos willingness toconsume the targeted foods (Busick Brooks Pernecky Dawson ampPetzoldt 2008 Worobey Ostapkovich Yudin amp Worobey 2010)

It would similarly be beneficial to establish the optimal lengthof the program and whether a program that runs for fewer than 4weeks is just as effective or whether a prolonged but less intenseprogram leads to longer-term changes in food liking Informal feed-back from the staff who participated in the activities suggests thattargeting the sensory activities around a single food each day ratherthan a single sense would result in less food waste and thereforereduce the cost of running the intervention

Finally it would be of both practical and theoretical interest toestablish the contribution of the activitiesrsquo involvement of differ-ent sensory modalities to the positive effects we have reported Whilethe current study has demonstrated the potential for a multisen-sory intervention to impact on childrenrsquos willingness to taste newfoods at mealtimes it cannot speak to the effect of each type of ac-tivity in isolation The participants in our study engaged in differentcombinations of sensory games depending on the days on whichthey attended nursery such that some children took part in gamesinvolving only two of the senses while others experienced activi-ties engaging all four targeted senses However the study was notpowered to investigate whether such differences affected theoutcome for individual children Nor do we know whether within

each sensory modality some activities exerted a greater influenceon childrenrsquos willingness to touch and taste the foods than othersA much larger study in which sub-groups of children were as-signed to different combinations of sensory activities would berequired to tease apart the relative impact of the different sensesin increasing childrenrsquos willingness to taste new foods and to designthe most efficient intervention with the greatest potential to enhancehealthy eating

References

Birch L (1999) Development of food preferences Annual Review of Nutrition 1941ndash62

Birch L Gunder L Grimm-Thomas K amp Laing D (1998) Infantsrsquo consumption ofa new food enhances acceptance of similar foods Appetite 30 283ndash295

Birch L amp Marlin D (1982) I donrsquot like it I never tried it Effects of exposure ontwo-year-old childrenrsquos food preferences Appetite 3 353ndash360

Bronstein P amp Crockett D (1976) Exposure to odor of food determines eatingpreferences of rat pups Behavioral Biology 18 387ndash392

Busick D B Brooks J Pernecky S Dawson R amp Petzoldt J (2008) Parent foodpurchases as a measure of exposure and preschool-aged childrenrsquos willingnessto identify and taste fruit and vegetables Appetite 51 468ndash473

Carruth B Ziegler P Gordon A amp Barr S (2004) Prevalence of picky eaters amonginfants and toddlers and their caregiversrsquo decisions about offering a new foodJournal of the American Dietetic Association 104 S57ndashS64

Cooke L (2007) The importance of exposure for healthy eating in childhoodA review Journal of Human Nutrition amp Dietetics 20 294ndash301

Dazeley P Houston-Price C amp Hill C (2012) Should healthy eating programmesincorporate interaction with foods in different sensory modalities A review ofthe evidence British Journal of Nutrition 108 769ndash777

Desor J Maller O amp Andrews K (1975) Ingestive responses of human newbornsto salty sour and bitter stimuli Journal of Comparative amp Physiological Psychology89 966ndash970

Desor J Maller O amp Turner R (1973) Taste in acceptance of sugars by humaninfants Journal of Comparative amp Physiological Psychology 3 496ndash501

Ellarsquos Kitchen (2011) Ellarsquos Explorers lthttpwwwellaskitchencoukare-you-a-nurserygt Last accessed 140312

Fedoroff I Polivy J amp Herman P (1997) The effect of pre-exposure to food cueson the eating behavior of restrained and unrestrained eaters Appetite 28 33ndash47

Gerrish C amp Mennella J (2001) Flavor variety enhances food acceptance informula-fed infants American Journal of Clinical Nutrition 73 1080ndash1085

Hagman U amp Algotson S (2000) Mat foumlr alla sinnen-sensorisk traumlning enligt SAPEREmetoden Stockholm Blomberg amp Jansson

Harris G (2008) Development of taste and food preferences in children CurrentOpinion in Clinical Nutrition amp Metabolic Care 11 315ndash319

Heath P (2013) Improving childrenrsquos responses to fruit and vegetables Picture-bookexposure and the impact of food familiarity and liking (PhD thesis) Universityof Reading

Heath P Houston-Price C amp Kennedy O B (2014) Letrsquos look at leeks Picture booksincrease toddlersrsquo willingness to look at taste and consume unfamiliar vegetablesFrontiers in Psychology 5(191) 1ndash11 doi103389fpsyg201400191

Hennessy M Smotherman W amp Levine S (1977) Early olfactory enrichmentenhances later consumption of novel substances Physiology amp Behaviour 19481ndash483

Houston-Price C Butler L amp Shiba P (2009) Visual exposure impacts on toddlersrsquowillingness to taste fruits and vegetables Appetite 53 450ndash453

Kalat J amp Rozin P (1973) ldquoLearned safetyrdquo as a mechanism in long-delay taste-aversion learning in rats Journal of Comparative amp Physiological Psychology 83198ndash207

Mustonen S Rantanen R amp Tuorila H (2009) Effect of sensory education on schoolchildrenrsquos food perception A 2-year follow-up study Food Quality amp Preference20 230ndash240

Mustonen S amp Tuorila H (2010) Sensory education decreases food neophobia scoreand encourages trying unfamiliar foods in 8ndash12-year-old children Food Qualityamp Preference 21 353ndash360

Organix (2010) Taste for Life lthttpwwworganixcomsupport-advicetaste-lifegtLast accessed 201010

Pliner P (1994) Development of measures of food neophobia in children Appetite23 147ndash163

Puisais J amp Pierre C (1987) Le goucirct de lrsquoenfant Paris FlammarionReverdy C Chesnel F Schlich P Koumlster E amp Lange C (2008) Effect of sensory

education on willingness to taste novel food in children Appetite 51 156ndash165Rozin P (1976) The selection of food by rats humans and other animals In J

Rosenblatt R Hinde C Beer amp E Shaw (Eds) Advances in the study of behaviour(Vol 6 pp 21ndash76) LondonNew York Academic Books

Thomas L (2007) Mange tout Teaching your children to love fruit and vegetableswithout tears London Penguin

Wardle J amp Cooke L (2008) Genetic and environmental determinants of childrenrsquosfood preferences British Journal of Nutrition 99 S15ndashS21

Wardle J Cooke L Gibson E L Sapochnik M Sheiham A amp Lawson M (2003a)Increasing childrenrsquos acceptance of vegetables A randomized trial of parent-ledexposure Appetite 40 155ndash162

5P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

support the potential for sensory activities to facilitate the intro-duction of healthy foods into childrenrsquos diets

The results of this study suggest several key avenues for furtherresearch First questions remain about the impact of the interven-tion beyond the taste test conducted We can say nothing about thechanges that might have occurred in childrenrsquos eating behavioroutside the nursery environment Future studies might explore theprogramrsquos impact beyond the confines of the nursery setting byasking parents about childrenrsquos eating behavior before and after anintervention at the childrenrsquos nursery Toddlers are likely to havemore influence over what they eat in the home environment thanat nursery and parents may notice if children ask for certain foodsat home or in the supermarket or are more willing to consume theseduring family mealtimes The effects of the intervention might alsobe specific to the foods targeted by the program or they might gen-eralize to other similar foods (Birch Gunder Grimm-Thomas amp Laing1998) or bring about positive changes in childrenrsquos attitudes towardnew foods in general Parents should therefore be asked to reporton childrenrsquos food neophobia andor food fussiness (Pliner 1994Wardle Guthrie Sanderson amp Rapoport 2001) as well as on theirconsumption of specific exposed and non-exposed foods before andafter a nursery intervention is conducted

The longevity of the effects seen in our lsquowillingness to tastersquo testalso remains to be established Previous work has questionedwhether a sense-based program for school-aged children can havea long-term impact on childrenrsquos eating behavior (Reverdy et al2008) In order to justify any recommendation to nurseries to im-plement a program similar to that used in our study we would needto know that the intervention is likely to cause lasting changes inchildrenrsquos willingness to consume the target foods Alternatively ifeffects were found to last for only 6 months or a year for exampleone might recommend repeating the program on a biannual orannual basis to keep childrenrsquos interest in the targeted foods lsquotoppeduprsquo

Other outstanding questions concern the locus of the positiveeffects of the intervention and how these might be optimized Wouldthe sense-based activities we employed have the same impact if theywere conducted in the home environment for example If parentsengaged in food-related activities with their child parentsrsquo own in-terest in and willingness to consume the fruits and vegetablesconcerned might be enhanced along with the childrsquos resulting inpositive changes to the parentrsquos food purchasing and eating behav-ior and hence an even greater impact on childrenrsquos willingness toconsume the targeted foods (Busick Brooks Pernecky Dawson ampPetzoldt 2008 Worobey Ostapkovich Yudin amp Worobey 2010)

It would similarly be beneficial to establish the optimal lengthof the program and whether a program that runs for fewer than 4weeks is just as effective or whether a prolonged but less intenseprogram leads to longer-term changes in food liking Informal feed-back from the staff who participated in the activities suggests thattargeting the sensory activities around a single food each day ratherthan a single sense would result in less food waste and thereforereduce the cost of running the intervention

Finally it would be of both practical and theoretical interest toestablish the contribution of the activitiesrsquo involvement of differ-ent sensory modalities to the positive effects we have reported Whilethe current study has demonstrated the potential for a multisen-sory intervention to impact on childrenrsquos willingness to taste newfoods at mealtimes it cannot speak to the effect of each type of ac-tivity in isolation The participants in our study engaged in differentcombinations of sensory games depending on the days on whichthey attended nursery such that some children took part in gamesinvolving only two of the senses while others experienced activi-ties engaging all four targeted senses However the study was notpowered to investigate whether such differences affected theoutcome for individual children Nor do we know whether within

each sensory modality some activities exerted a greater influenceon childrenrsquos willingness to touch and taste the foods than othersA much larger study in which sub-groups of children were as-signed to different combinations of sensory activities would berequired to tease apart the relative impact of the different sensesin increasing childrenrsquos willingness to taste new foods and to designthe most efficient intervention with the greatest potential to enhancehealthy eating

References

Birch L (1999) Development of food preferences Annual Review of Nutrition 1941ndash62

Birch L Gunder L Grimm-Thomas K amp Laing D (1998) Infantsrsquo consumption ofa new food enhances acceptance of similar foods Appetite 30 283ndash295

Birch L amp Marlin D (1982) I donrsquot like it I never tried it Effects of exposure ontwo-year-old childrenrsquos food preferences Appetite 3 353ndash360

Bronstein P amp Crockett D (1976) Exposure to odor of food determines eatingpreferences of rat pups Behavioral Biology 18 387ndash392

Busick D B Brooks J Pernecky S Dawson R amp Petzoldt J (2008) Parent foodpurchases as a measure of exposure and preschool-aged childrenrsquos willingnessto identify and taste fruit and vegetables Appetite 51 468ndash473

Carruth B Ziegler P Gordon A amp Barr S (2004) Prevalence of picky eaters amonginfants and toddlers and their caregiversrsquo decisions about offering a new foodJournal of the American Dietetic Association 104 S57ndashS64

Cooke L (2007) The importance of exposure for healthy eating in childhoodA review Journal of Human Nutrition amp Dietetics 20 294ndash301

Dazeley P Houston-Price C amp Hill C (2012) Should healthy eating programmesincorporate interaction with foods in different sensory modalities A review ofthe evidence British Journal of Nutrition 108 769ndash777

Desor J Maller O amp Andrews K (1975) Ingestive responses of human newbornsto salty sour and bitter stimuli Journal of Comparative amp Physiological Psychology89 966ndash970

Desor J Maller O amp Turner R (1973) Taste in acceptance of sugars by humaninfants Journal of Comparative amp Physiological Psychology 3 496ndash501

Ellarsquos Kitchen (2011) Ellarsquos Explorers lthttpwwwellaskitchencoukare-you-a-nurserygt Last accessed 140312

Fedoroff I Polivy J amp Herman P (1997) The effect of pre-exposure to food cueson the eating behavior of restrained and unrestrained eaters Appetite 28 33ndash47

Gerrish C amp Mennella J (2001) Flavor variety enhances food acceptance informula-fed infants American Journal of Clinical Nutrition 73 1080ndash1085

Hagman U amp Algotson S (2000) Mat foumlr alla sinnen-sensorisk traumlning enligt SAPEREmetoden Stockholm Blomberg amp Jansson

Harris G (2008) Development of taste and food preferences in children CurrentOpinion in Clinical Nutrition amp Metabolic Care 11 315ndash319

Heath P (2013) Improving childrenrsquos responses to fruit and vegetables Picture-bookexposure and the impact of food familiarity and liking (PhD thesis) Universityof Reading

Heath P Houston-Price C amp Kennedy O B (2014) Letrsquos look at leeks Picture booksincrease toddlersrsquo willingness to look at taste and consume unfamiliar vegetablesFrontiers in Psychology 5(191) 1ndash11 doi103389fpsyg201400191

Hennessy M Smotherman W amp Levine S (1977) Early olfactory enrichmentenhances later consumption of novel substances Physiology amp Behaviour 19481ndash483

Houston-Price C Butler L amp Shiba P (2009) Visual exposure impacts on toddlersrsquowillingness to taste fruits and vegetables Appetite 53 450ndash453

Kalat J amp Rozin P (1973) ldquoLearned safetyrdquo as a mechanism in long-delay taste-aversion learning in rats Journal of Comparative amp Physiological Psychology 83198ndash207

Mustonen S Rantanen R amp Tuorila H (2009) Effect of sensory education on schoolchildrenrsquos food perception A 2-year follow-up study Food Quality amp Preference20 230ndash240

Mustonen S amp Tuorila H (2010) Sensory education decreases food neophobia scoreand encourages trying unfamiliar foods in 8ndash12-year-old children Food Qualityamp Preference 21 353ndash360

Organix (2010) Taste for Life lthttpwwworganixcomsupport-advicetaste-lifegtLast accessed 201010

Pliner P (1994) Development of measures of food neophobia in children Appetite23 147ndash163

Puisais J amp Pierre C (1987) Le goucirct de lrsquoenfant Paris FlammarionReverdy C Chesnel F Schlich P Koumlster E amp Lange C (2008) Effect of sensory

education on willingness to taste novel food in children Appetite 51 156ndash165Rozin P (1976) The selection of food by rats humans and other animals In J

Rosenblatt R Hinde C Beer amp E Shaw (Eds) Advances in the study of behaviour(Vol 6 pp 21ndash76) LondonNew York Academic Books

Thomas L (2007) Mange tout Teaching your children to love fruit and vegetableswithout tears London Penguin

Wardle J amp Cooke L (2008) Genetic and environmental determinants of childrenrsquosfood preferences British Journal of Nutrition 99 S15ndashS21

Wardle J Cooke L Gibson E L Sapochnik M Sheiham A amp Lawson M (2003a)Increasing childrenrsquos acceptance of vegetables A randomized trial of parent-ledexposure Appetite 40 155ndash162

5P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6

Wardle J Guthrie C A Sanderson S amp Rapoport L (2001) Development of thechildrenrsquos eating behaviour questionnaire Journal of Child Psychology amp Psychiatry42 963ndash970

Wardle J Herrera M Cooke L amp Gibson E (2003b) Modifying childrenrsquos foodpreferences The effects of exposure and reward on acceptance of an unfamiliarvegetable European Journal of Clinical Nutrition 57 341ndash348

Whichelow M amp Prevost A (1996) Dietary patterns and their associations withdemographic lifestyle and health variables in a random sample of British adultsBritish Journal of Nutrition 76 17ndash30

Worobey H Ostapkovich K Yudin K amp Worobey J (2010) Trying versus likingfruits and vegetables Correspondence between mothers and pre-schoolersEcology of Food amp Nutrition 49 87ndash97

Appendix List of activities used to explore foodsrsquosensory properties

Sight1 Compare the difference between the outside of the whole food

and the inside of the food after it has been cut open

2 Find items around the nursery that match the color of each food3 Draw a picture of each food

Sound1 Sing a nursery rhyme about fruit and vegetables (song sheet pro-

vided adapted from Old McDonald Had a Farm)2 Run toward a picture of the food when its name is called3 Listen to the noise made when the food is snapped or squeezed

Touch1 Feel the texture of the outside of the whole foods2 Feel the texture of the foods after theyrsquove been chopped up3 Compare the texture of the chopped up foods before and after

they have been cookedSmell

1 Listen to a story about a child smelling foods in the shop andat home smelling the foods while listening to the story

2 Smell the cooked food3 Compare the smells of the cooked and uncooked food

6 P Dazeley C Houston-PriceAppetite 84 (2015) 1ndash6