Sensory Evaluation Summary PDF

introduction Origin Challenges Sensoryevaluation sensory method actual Afterfoodrestriction qualitymoreimportantthan quantity session to assess prod foodpositing competition sensoryana is a centraltool Sensoryand...

introduction Origin Challenges Sensoryevaluation sensory method actual Afterfoodrestriction qualitymoreimportantthan quantity session to assess prod foodpositing competition sensoryana is a centraltool Sensoryand aspecific objforthestudy Challenge climtechang 2.5 5.2c 9 a hypothesis statistical ans of thesenory Population evaluation test Definition Sensory and a scientificdisciplineused toevoke measure analyse interpret reaction tothese characteristic offood astheyare perceivedbythesenses ofsight smell taste touch hearing Decisionmakingprocess Change formpureinstinctprocess mix of sensorial conscious Rotional emotional Rational decision process w currentmisfires resulting in obesity Emotion early instinct served as evolutionary tool givemore rational thought through natural selection nowtheyare interweaved choose options that emotion Orbitofrontal cortex 1OFC Gutinstint Enteric Nervous System IENS Senses food characteristics Appearance flavour Aroma texture sound no chemical composition Basictaste Bitterness saltiness sourness sweetness umami Newtaste fat starchy Human sense gut feeling taste hearing smell inot Extra sensory Sight touch perception ESP Complexity hotchotaste better in orange cup Foodtaste better using heavy cutlery SenseStimulating Cutlery Commontest 1 DifferenceTest detectsmall diff in food discrimination test 2 Affectivetest consumer Acceptance test to other focus on 1 attribute like or dislike no compare Preference test rank allsample all attributes 3 Descriptivetest describe theperceivedsensory qualitycontrol trained panellist Basic Concept Basicsenses sensory perception Color cones 1Blue green red Rod 1White Black Munsellcolor system Hue Differentiate color parometer  Value lightness Chroma colorpurity VisionProblems Red Green colour deficiency Blue Yellow color Complete color Ishihara test detect color inred green the type deutan or proton texture use asenses foodquality can change during storage bymoist T Divided in Tactile oral Mouthfeel or trigeminalsensation phy chem in mouth if Ftext Indirect touch eg useutensil Visual perceived beforetouching Auditive Noise eardrum wore auditory nerve Signal o estiem in lip Bite mastication saliva pipettp.int ifs tjf asP smtttgf.tt muthfeelechemrunw saliva Odour Olfactory nasal cavity perception of odour aroma Orthonasal pathy way through nose reacholfactory bulb Aroma fromthemouth through Retronosal pathway odour vs aroma salivi's nitrates Gustatory sense food Dissolved b y thetastebud signal to dendrites to brain Sensation threshold Recognitionthreshold flavour basistastes oromo trigeminal sensation mouthfeel Mouthfeel trigeminalsensation trigeminal nerve manifest as aggressive or irritating sensation Astringent tannins cold heat it or chemi metallic substances Pungent chilli Bubblyor fizzy carbonic a Sensory Bases Psychological factors Distraction Error various stimuli outside thetest Solution quietarea eles devices Habituation Error evaluate similar on regular basis Tintinlusiarytorunsample Solution introduce adulterated samples wmoiicinunoaahsosldliiv.s.ms Stimulus logical Error judment an irrelevantstimuli or visal some presentation color size form HaloEffect 1 attribute isevaluated affect each other Solution evaluate in oder limit attribute OrderEffect influ by previous sample 1stsample gethigher score I CentralTendency Error usemiddle sample Solution randomise theorder for affective testuse a dummysampleas7  Contrast Convergance error contrasteffect 2 sample are very diff exaggeratetheans Convergence effect similarprd w highvariation evaluate w lower diff than they have Solution Randomise order eliminating sample that are very diff Culturalfactor Cultural or Geographical Origin diff in terminology Restriction on some prd variation in familiarity w someattributes Solution clarifythetranslation terminologyused Be aware of culture trends Collectinfo based on cultural norms Conduct a filter w participant regarding anyfood restriction Physiological factor Adaptation Continous exposure b sensitivity Solution limit num of sample ensuretime btw sample provid palate cleanser Breaktime Percetual interactions high I first affect next sample Solution experimental design to mini unwanted interactions Physical condition health no disorders med age stress t of theday Solution filter Q Instruct participant not to consume food leaf 1 hr allsessions in a similarschedule monitor participant performance for Possible change Discrimination Tests discrimination test objective method testhowsigare the perceived diff inprd aim assess whether a sensory difference exits btwsample Sensitivetest use assessrelativeintensity ofspecific compounds related to odour flavour Selection of panellist Threshold identification Determine tastingability Test level of sensitive of panellist Thesholdtest  Absolute lowest can detect Recognition n to recognise compound Differential Idetect differences or changes Terminal highest candetect an increase Differencetest use possibilityto detect any diff btw sample diffare veryslight most useful sensitive sensory test Basedon forced decision require Trained panel Num 20 50 Conduct Poweranalysis notsuit for lingering aftertaste or strong flavours for Quality control Panelist selection trainning Benchmarking new prd Triangletest use control quality Shelflife studies Prod improvement Addition or substitution ofingre replace a pud consregonise Selection training of panellist require may reuse panellists once notadd panellistafterdata has been an 3 samplepresented thesame time Info to panellis 2 same are the same 7 is different taste from left to right Cleansepalate btw sample identify the diff sample Sample tasted leasttwice every intensive evaluation Testdesign constant reference set 1 sample as different throughout theevaluation need 7 set of codes use when I sample is limited 3 possible order Balanced ref diffsample splitevenly 2 complete set of codes to bias 6 possible order label 3 digitrandom codes Sample order of presentation Balanced order allpossible sample orderareusethesomenumb oftime bestwhen panellists is a multiple of it Randomorder all panellist havethe somechance of getting any ofthe possibleorderof presentation 1 3 ofpanellist mustselectthecorrectans sig difference  calculation Duo Triotest use identify diff Quality control when reference are well known NPD Change in pud technologies Selection training Shelf life studies Change in prd formulation require taster triangletest sample presented thesametime Sample in predetermined random order Info sample labelled as R is the ref 2 sample are assigned a 3 digitrandom code Taste R first Then taste from left toright Identifythe samplediff fromR Sample maybetasted twice Testdesign constant net set 1 sample as R I set of code for limitedquantities trainning 2 possible order Balanced ref R can be either testsample 2 complete set of code fossible order Sample of presentation to fast Inder need thenellist less efficientthantriangletest 50 correct are 50 of panellist must selectthe comet ans Sig Calculation Forced choicetest 1 Paired comparison tests Earliest discrination test Simplest sensory test Minimise participant Fatigue  Panellistare asked to assess a specific criterion Type 1 Difference Prine use comparing prd w homogenois Fiiity but heteljeiissitb.to Training control of panelist improve prd when diff is unknown require 2 sample presented sometime predetermined randomorder info sample are assigned a 3digit left righ respond if sample arethe same or diff tasted once 4 possibleorder 2 I Mdetermine diff level direction of the diff selectby differential threshold Testsensitivity ofpanellist improve pd require sample time fame f left right select a sample based on specific attriblee sweeter tastemore than once but in some order Sampletent beidentical except interested attribute 2 possible order 50 chance of correctone 2 n Alternative forced choice on AFC Similar to directional but w 3 sample Only 1 sample diff sample must be identical A not A test willnot give info abt the specific diff use have slight variations in visual attributes when sypple NPD prototype closed to ref require panellist familiar w control 2 5 samples pif know no A info single sample is present as A cutenthey not familiar w A remove 1 sample present 2ⁿᵈ sample 1 by1 Ask which one is similar to A Step repeated w all sample one at a time  Stat Chi squared test Poweranalysis Probability of detecting a diff assess number of panel are enough to find sig diff Consumer Test participant regularconsumer no experience in sensory evaluation measure consacceptance preference feeling emotions usage perception Require focusgroup seveal session of 6 12 participant Regular consumer test 40 participant largescale or market test 100s 1000s sideal 700300 participants1 city number of participant depend on num of sample amount of sample available expected mean value expected SD Poweranalysis 7 wayANOVA useafter or before location Sensory laboratory central location Rented facilities Retailing location supermarket At home realenvi opt sample Questionnairesip minimum siipiet.to iiafoiiie Psample question Selectioncons filter Q target participantmust bepredefined Qualitative test Focus groups Assess can attitudes toward prd  Non numerical data Open Q discusion Interaction group interviews w a moderator use Brainstorming lexicongeneration Questionnaire development NPD sprdideas exploreprototypes prd use tonumber of pototype knowwhatcans lookfor Consumptionattitudes Adv con may rise issues flexible Discussion In depth understanding Dis may createbias Moderator Severalsession are required Time consuming 21hr1 High expenses ilocation incentivel Process prol definition 7Th géF s numofsession participant select a moderator design a discussionguide recruitparticipant to conductthestudy dataand Report Quantitativetest Numerical data Subjective response Individualtest TYPE Acceptance test determine sample is more acceptable like thanothers if I intensities of specificattributes based on liking Type of interval scale 9point hedonic scale Hedonic continuous scale 3 point s point JARscale 11 Hedonic test level of pleasure sensations elicited fromdifferent stimuli IEP tingeof c attribute from visual odour mouthfeel tastes overallliving use NPD narrow num of protypes prd release 2 sample Sample presentation monadic one at a time Predetermined random order Randomised complete block design participant follow the Q roting the prd attribute Stat ANOVA posthoc test Multivariate data and 1.2Justaboutright IJARItest Assess desirability of aspecificattribute use determin optimum level ofattribute ofinterest comparison of prototype tofind most optimal attribute levels Adv test a single prd combined w hedonic test Dis cons may not be familiar w attribute use common wellknown attri i Reformulation for a singleattribute mayaffect theprd need retestingafte reform Result compare JAR value assess the closest to ideal value 50 1 2 Preference test if ii ieference based on overall liking specific attribute liking T paired preference Similar to paired comparison discriminativetests use NPD reformulation Benchmarking QC Adv easy to conduct Nofatigue easy vapid ana not requirerepetitions Dis few info highprobability of error unknown reason require 2 sample simultaneously participant must select the preferred sample Ana 2 tailed binomial table for paired pref 2.2Ranking Rann order prd based onpreference use NPD change sup ingre reform Benchmarking Adv easy rapid ana easy to conduct Additional data when conducted w acceptanie test Dis Fatigue when 9num of sample when ranting pref for several attributes in session  require 3 sample Balanced random order maytasted 1 if more than 7 attribute new set of sample ediff label rank form most o least preference Ano assign a num of each ranked sample 17 least n most sum value Non parametric data and 3 Check allthat apply ICATAITest assess a w multiple options lexicons Terms Adv simple versatile Suitable for larger set ofsample Dis sonmaynot use deep cognitive processing toaccurately respond require monodic presention w Bolonce randomorder select all option that pply iset mini on maxi num if term to select Data frequencies Ana Descriptivetest provide a description of a prd Integral sensory description considering allsensations to evaluate prd qualitative quantitative sensoryaspects mayuse a singleprd or comparison of pro prototypes use NPD givethe idea ofsensory profile Shelf life cin terms of taste Consumer complaint diff fromusually generate descriptors Assess panelist performance QA ctostethesame assample Training 150Regulations o fixed panel for diffdescriptivetest Vision Ishiharatest Color intensity ranting ifnotposs can be ponelistwhen notenough pp removefrom visualtest can'ttaste Taste Taste detection threshold Ageusion  Acuity discrimination ability detect identifytaste Matchingtest 80 correct can'tsmell Odour aroma Detection threshold Anosmia Acuity discrimination ability Descriptive ability imeasured foodprd identifyaroma Texture Viscosity Carbonation related terms Discrimination ability Descriptive ability Quantitative Descriptive Analysis IQDA Panelist 10 16 ideally 12 Scale is cm non structured cantscale Training time 1ˢᵗ 3 week 190min week Existing 1 90minsession Panellistrequiredto develop language descriptor of prod training Rate theintensity ofeach descriptor sensory session Can combine w 180 to assess proticipel Adv reproducibility Consider all attributes Qualitative descriptors quantitative intensities dota Dis highcost Training may not enough Data ana ANOVA spider charts Multivariatedata and Principal component ana PCA Multiplefactor and MFAI Spectrum Descriptive Analysis Panelist 15 Scale Absolute scale shire ref 10 or 15 cm unstructured Training time 6 8 h week for 14 week Panellist required to use of specific ref to helpdescribing prd rote intensity of descriptors Adv Intensive prescreening eliminate panellist Reproducility Dis Time consuming high cost Data ana ANOVA spider charts Multivariatedata and Principal component ana PCA Multiplefactor and MFA Histogram for panel performance flavour Profile 1ˢᵗdescriptive method Panellist 45 Scale s level Notpresent Threshold Slight Moderate Strong Adapted larger scales Panellistrequired to extensive training generate flavour descriptor bymoderator Rotetheintensitylevel obtain a single overall assessment Qualitative spiderchart or histogram Adv highprobability of consideringallpossible descriptors May use ref Dis high cost Time consuming Qualitative 1Nostatanal limited to flavar Scale not sensitive Texture Profile Panellist 6 9 Scale lengthvaries depending ondescriptor Panellist required to extensive training Generate descriptorbased on ref Rate based on netlevel Adv full description Quantitive Dis Time Cost focused only on Texture use of diff scale no specific scale confuse Ana ANOVA spider charts Multivariatedata and Principal component ana PCA Multiplefactor and MFAI Timeintensity 12 14 panellist Scale 10 or is cm unstructured Panellist require to generate descriptor Rate intensity various time intervals chewinggum howlongmintstay Adv detailed info on change in flavour aromas texture over time Dis descriptor to rote overtime depends on the level of training ofpanellists Ana Meandata plotted in one cintensityovertime Calculation Imax Timedecay Rapid method 1 Free choice 10 20 p Individual ans 6 levelscale Generalized Procrustes ano  2 Flash profile 10 20 p Ranking 11 point scale Generalized Procrustes ano 3 Napping cons or trained 2Dmapbased on similarity Descriptor ofeachsample Frequencies Correspondence ana multiple factor and 4 Sorting 20 p Group sample bysimilarity factspIlnc iigthe sample i.EE Statistic Acceptancetest 1 Hedonictest ANOVA rating of 2 samples Assess sig diff btw a group of samplefor eachattribute Mayana only 1 attribute time At α 0.05 p value α sig diff ANOVA notassess which one are diff A posthoc meanseparationtest Meanseparationtest fishers leastsignificant difference LSD test Duncan test way to report ANOVA Bar chart spider chart Tukey test Software Excel XLSTAT SAS Case study 2 Just About Right ITARITests Bipolarscale Anchored bothsidesby opposite point Ana Frequency of selection Penalty and JAR hedonictests Howmuch does a specific attribute affecttheoverall liking Casestudy Preferencetests 1 Paired preference test Count number as most preferred 2 tailed binomial table 2 Ranking Friedman test lormond 19777 table 1 least preferred n most preferred Checkallthatapply CATA test 1 Frequency of selection easy but limited info used for further and stelloptionw mostselectedbutnottell sigi Cochron Qtest Multivariated data and relationship 2 Multivariate data and notthemost used in CATAI Relationship btw variable emotions association btw sample variable lemos Descriptive quantitative test spiderchart meanvalue ofintensitiesofdiffattributes  2 Multivariate dataana PCA optional usewhen having multiple variable w quantitative data Assess interaction association ofsample variables and multiple attribute a time find relationship or interactions developed using the mean values persample variable Facilities Samplepreparation Sensory facilities 7testing areas net Each room conforming thesensorylob must have fit focus group T RH control airflow fromeachareaone mus beindependent to avoid contamination undesirable odour Individual booth room Booth 2 Preparation Area Servery on needs giant kitchen 3 Office  pppl lif in servery orhitchen usenow 5 Waiting room Chairs 12 Sample preparation 1 Serving T 2 Preparation 3 Serving size 4 Carriers prd that consume w other notstrongflavour Sample evaluation 1 Number of sample 2 Palate cleanser refresh sensation New EmergingTechnologies in Sensory Evaluation Consumer Sensory Evaluation our response to foed classified unconscious s conscious 1 Traditional method conscious Acceptance test preferencetest  2 Non Invasive Biometrics computer Vision Unconscious 3 Machine learning modelling unconscious Conscious Biometrics Identify or recognize physiological behavioural distinctive characteristics 1 Autonomic Nervous System heart rate pupil Dilation Body T Respiration 2 Emotion 3 Bio Sensory App 4 Method to measure Biometrics emotional response heat rate body T voice analysis aeIpffetePiiiad p Novel non invasive method TPP'jÉ if cons awareness Sig effect on experience stress Photoplethysmography PPG measure 02 lightreflected byblood  2 Potypia Novel non invasive method itend to have lower T surface may Ift deny 3 Emotion 4 EyeTraching measure gaze movement positioning 5 Brainwave Electroencephalogram IEEG brainactivity Machine leaving Modelling Computer based system trained to find patterns among a dataset to classify or predictspecific Piaget.pe aiiinte'ijfeiiihneimasiineno'intersenseagi.cat iismision.robotic Machine learning EE IE hsupervied supervised ir mEtumesi Tetesointo Clustering Regression assing diffcompound ofnympygal Classification classbtw diff brand Association type ofprd other Emerging method  case studies BeerTasting Been EyeTracking EyeTracking label Sensory in Products Development 1 Necessity market Research Benchmarking Trend competitor soles 2 Idea Concept focus group New totheworld New tech ingre concept lineextensions Newflavor color Healthier Metoo copies competitor 3 Formulation Ingre Basic flavor Additive Concentration Physicoinemical ana pH Acidity Totalslouble solids 4 Screening Goal verify thedesired prd tonum of prototypes mox 6 sample it compel Differences test line extension for reformultion Internal consumer test Acceptance test Hedonic JARP Employee Prototype vs competitor Nosolify in minimum acceptance score 7 Reformulation 5 Product Selection  Consumer test Acceptance test Hedonic JAR dohedonic I 2 sat do preference Preference test Ranking CATP prototype 1 31 competitors close iritis select by cost comercail o Packaging Primary Extend shelflife Secondary Itection Tertiary Flute Resistance ECT BCT I 8 Technical sheet Include 9 First Batch Production Descriptive sensory profile Physicochemical and lose stole us top Microbiological any Proximate ana composition offood for lobel Shelf life Verification of process food safety system 10 Shelflife Accelerated shelf life from ist batch Incubator JH Testing sampling times depend onprd Physicochemical ans aw PeroxidesAcidity Microbiological ans Pathogen Mould yeast Sensory ans Descriptive test compare w controll Calculation Arrhenius equation 1ˢᵗorder kinetics equation Q101 Subtract 10 fromthe shelflife estimation 11 Quality Assurance monitoring compare w standard sensory phychem micro Case study Question from prep Terminalsheet First batchproduction createsensory profile Prodselector ctaceep.CATPI tastedetectionthresholdcageusia  mouthfeel chilli sort 2 80 Eye tracker measure detect gaze eye pupil movement positioning can use to study about where consumer Theprincipal component analysis PCA shows lookingon packing what is themostimportant herelationship btw s sample FS head movement or interest on packing the fist second principal component epc epc is 74.23 of the variability in thedata  Serving T T no 120 25 c Preparation give a knife to participant to spread in a bread or cracker as a carriers sensitivitytest Differential test Serving size 28 i c detect diff Number of sample g maximum 6 sample but this strong prod 4 samples Palate cleanser warm water or chamomiletea Rapid method 1 Free schoice 2 Floch profile 3 Napping 4Sorting Cinna HT got the highest pe gyptAftertne no sig lettler face scale be some birds might be able to understand theword PCI PC2 84 52 Positiveteen NegativePCT PC 1 per