CoffeariumBeta
  1. Plant
  2. Farm
  3. Process
  4. Trade
  5. Roast
  6. Brew
  7. Cup
Science · Tasting

Crossmodal effects

How one sense changes what another reports – why the colour, shape and feel of a cup, a label, a sound or a smell can shift how sweet, sour or good the same coffee seems, and why cuppers taste blind.

Across the chain: Science

Source-checkedby Coffearium verification pass· Last source-checked 21 Oct 2026· Prepared with AI assistance· 19 sources (16 tier A)· Updated 21 Oct 2026
Reading depth
Contents
  • Crossmodal effects: one sense changes what another reports. In tasting studies, the same coffee was rated differently when only the cup or a label changed.8,10,16
  • In a 2014 experiment with 36 participants, a café latte was rated less sweet from a white mug than from a blue or a transparent one.8
  • In a 2020 study, 81 Q graders rated the aftertaste of one coffee as drier when they drank it from a rough cup rather than a smooth one.10
  • The coffee cup studies had 18–457 participants per experiment or study and all share one co-author; no independent replication was found for this entry.8,9,10,11
  • The Specialty Coffee Association (SCA) requires identical cups for cupping and keeps information about a coffee away from its tasters to avoid bias.1,2

One sense changes another#

Flavour is built from several senses at once. Taste, smell and touch act together and interact to form one perception.3 A 2015 review calls flavour perception perhaps the most multisensory of everyday experiences.4 How the senses combine is explained in the entry on how we taste coffee.

A crossmodal effect goes one step further. A cue in one sense changes what a person reports in another. The cue may belong to the coffee itself, such as its smell. It may also come from outside the cup: the colour of a mug, a word on a label or the music in a room.15

The SCA calls attributes such as origin, processing information or certifications extrinsic. They cannot be discovered from the coffee itself.2 According to the SCA, extensive research has shown that extrinsic attributes affect both sensory perception and the impression of quality.2

Smell and taste in the cup#

The closest crossmodal effects happen inside the mouth. In a 1999 model study, odours added to sugar solutions changed how sweet they tasted. Some odours raised the sweetness; others lowered it.6 The more sweet an odour smelled, the more it raised the tasted sweetness.6 That study used solutions, not coffee.

In espresso, the effect has been shown in the other direction. Sugar or a sweetener raised the caramel ratings of the aroma and lowered roasty ones.7 The sweeteners are not volatile, so a taste changed how an odour was perceived.7 As of June 2024, the SCA was researching whether coffee’s own sweetness is a taste or a crossmodal effect of sweet-smelling compounds.2

Temperature works partly through physics. In a 2017 study, one coffee released more volatile compounds when served hotter.19 Serving temperature has its own entry.

The cup: colour, shape and texture#

The cup is the cue studied most often in coffee. Three kinds of change have been tested.

  • Colour. In a 2014 study, a café latte was rated less sweet from a white mug than from a blue or a transparent one.8 In a 2019 study with 457 participants, cup colour shifted expected and rated sweetness and acidity of specialty coffee.11
  • Shape. In a 2018 study at a coffee event in Brazil, the same coffee smelled stronger from a tulip-shaped cup. It tasted sweeter and more acidic from an angular “split” cup.9
  • Texture. In a 2020 study, coffee drunk from a rough cup was rated drier in aftertaste than from a smooth cup. Both Q graders and amateurs reported this.10

In these studies, the coffee in every cup was the same. Only the vessel changed.8,9,10 Professionals were not immune. In the shape study, both experts and amateurs found the aroma strongest in the tulip cup.9 The design and history of cups are covered in the entry on cups and vessels.

Words, labels and the look of the drink#

What people are told also changes what they taste. In a 2013 study, participants tasted two cups and were told that one held “eco-friendly” coffee. Both cups held identical coffee.16 Participants preferred the taste of the labelled cup and would pay more for it – at least those with strong attitudes towards sustainable consumption.16

Specialty coffee shows the same pattern. Information about a coffee changed which flavour notes consumers ticked, and the word “fermentation” on a label made one coffee seem more acidic.17,18 These label studies are described in the entry on flavour language.

The look of the drink counts too. In a 2015 study, latte art raised the value people expected from a milk coffee. They were willing to pay more for it than for the same drink without art.14

Room, sound and light#

For the setting around the cup, a 2019 review gathers the coffee evidence. It finds that visual, auditory, olfactory and tactile aspects of the environment all influence the coffee experience.15 Two of the studies it reports, on lighting and on music, were never published.15

Two examples show the state of the evidence. In an unpublished laboratory study, people who liked weaker coffee drank more of it under dim light.15 In an unpublished study with seven tasters, the results suggested that coffee flavour, bitterness and astringency seemed less intense with jazz.15 Such results are leads, not established facts.

Why cuppers taste blind#

Cupping protocols hold many of these cues constant. SCA Standard 102-2024 requires cupping vessels of tempered glass or ceramic. All cups in a session must have identical volume, dimensions and material.1 The cup studies suggest that a different cup could shift ratings.8,9,10

Information is kept away from the tasters as well. The SCA states that tasters can be easily biased by extrinsic information about a coffee. It therefore keeps the physical, sensory and extrinsic assessments separate.2 According to a sensory textbook chapter, samples should be coded and served in random or balanced order.3 The methods are explained under sensory science methods.

Reference values#

Study Participants Cue changed Reported effect Limits
Van Doorn et al. 2014, experiment 18 18 volunteers, 6 per mug White porcelain, blue porcelain or glass mug; café latte White mug: higher flavour intensity than glass Small sample and mugs differed slightly, stated by the authors; no effect sizes
Van Doorn et al. 2014, experiment 28 36 volunteers, 12 per mug Glass mug with white or blue grip, or plain glass White: less sweet than blue and glass Only the grip colour differed; no effect sizes
Carvalho and Spence 20189 276 analysed (151 professionals, 125 amateurs) Tulip, open or split white cup, 173.0–176.5 g Aroma strongest in tulip; sweetness and acidity highest in split; split liked least by amateurs Expectations not measured; no effect sizes
Carvalho et al. 202010 81 Q graders; 117 amateurs Smooth porcelain (173.0 g) or rough ceramic cup (142.1 g) Aftertaste drier from rough cup in both groups Each group tasted a different coffee; cups also differed in shape
Carvalho and Spence 201911 457 in three experiments White, pink, yellow or green cup; sweet and acidic coffees Colour shifted expected and rated sweetness and acidity; liking fell when colour and coffee did not match Abstract only read for this entry
Van Doorn et al. 201713 309 online (China, Colombia, UK) Pictures of eight empty mugs Narrow mugs: more aroma expected; short mugs: more bitterness and intensity expected No coffee tasted
Sörqvist et al. 201316 Three experiments “Eco-friendly” label on one of two identical coffees Labelled cup preferred and valued higher by those with strong sustainability attitudes Abstract only read for this entry
Gater 2010, in a review15 7 No music, jazz, classical or rock Jazz: flavour, bitterness and astringency rated less intense Unpublished; trends only

The table lists coffee studies opened for this entry. Participant numbers are those analysed, where stated.

How solid the evidence is#

The coffee studies are few and close together. All six peer-reviewed cup and latte art studies opened for this entry share one Oxford-based co-author.8,9,10,11,13,14 No independent replication of a coffee cup effect was found for this entry. Most effects are reported as significance tests on 0–10 or 0–100 line scales.8,9,10 The 2014 colour study and the 2018 shape study report no effect sizes.8,9

0100200300400500participantsMusic, unpublished(Gater 2010, in a review)715 Mug colour, experiment 1(Van Doorn et al. 2014)188 Mug colour, experiment 2(Van Doorn et al. 2014)368 Cup texture, Q graders(Carvalho et al. 2020)8110 Cup texture, amateurs(Carvalho et al. 2020)11710 Cup shape(Carvalho and Spence 2018)2769 Mug shape, pictures only,online(Van Doorn et al. 2017)309, no tasting13 Cup colour, threeexperiments(Carvalho and Spence 2019)45711
Data table
participantsSource
Music, unpublished (Gater 2010, in a review)715
Mug colour, experiment 1 (Van Doorn et al. 2014)188
Mug colour, experiment 2 (Van Doorn et al. 2014)368
Cup texture, Q graders (Carvalho et al. 2020)8110
Cup texture, amateurs (Carvalho et al. 2020)11710
Cup shape (Carvalho and Spence 2018)2769
Mug shape, pictures only, online (Van Doorn et al. 2017)309, no tasting13
Cup colour, three experiments (Carvalho and Spence 2019)45711
Participants in the coffee studies opened for this entry, as reported (analysed numbers where stated). The online study showed pictures only; the music study was unpublished.8,9,10,11,13,15

Several effects are narrow. In the 2014 colour study, each experiment changed only one of six rated attributes.8 The intensity effect of experiment 1 did not reach significance in experiment 2, where only a trend was seen.8 In the 2020 texture study, the cup changed acidity for Q graders but sweetness for amateurs. Body did not change in either group.10

Designs also differ in what they measure. Online studies with pictures of empty mugs measure expectations, not perception.13 The 2018 shape study measured perception but not expectations.9 The cups in the texture study also differed in shape and in weight, by about 31 g (our calculation from the reported 173.0 g and 142.1 g).10 A 2017 review of receptacle shape notes that weight and similar factors need to be controlled. It adds that the science that explains such results is not there yet.12

CueWhat tasters reported1Smellsweet odour in sugar water 6 2Tastesugar or sweetener in espresso 7 3Cup colourlatte, specialty coffee 8,11 4Cup shapetasted 9 · pictures only 13 5Cup texturesmooth or rough 10 6Latte artmilk coffee 14 7Labels, informationeco-label, fermentation 16,18 8Music, soundunpublished 15 9Lightingunpublished 15 Aroma notes, aroma strengthSweetnessFlavour intensityAcidityAftertaste drynessBitternessLiking, expected valueAmount drunk
Which cue has been linked to which coffee judgement in the studies opened for this entry. Solid arrows: a published study in which coffee was tasted. Dashed arrows: a model solution, expectations only, or an unpublished study. Arrows show reported directions of influence, not their size. Small numbers on the cue labels are sources. Schematic.6,7,8,9,10,11,13,14,15,16,18

Proposed mechanisms#

Physical change. A cup can change what reaches the nose. Glass shape can change the headspace above a drink.12 The review notes, however, that such physico-chemical differences do not always lead to perceptible differences. Claims that a shape changes how liquid flows over the tongue are rarely backed by evidence.12 Where only the outside of a vessel varied, physical accounts appear implausible.12 In the second experiment of the 2014 mug study, only the colour of the grip differed.8

Expectations. Cues set up an expectation before the first sip. Drinks are rated more favourably in vessels judged congruent with their contents.12 In the 2019 colour study, liking fell in incongruent pairings of cup colour and coffee.11 Information can change expectations and, through them, sensory perception.17

Crossmodal correspondences. People match features across senses in consistent ways. According to the shape review, rounder forms are associated with sweetness and fruitiness, angular forms with bitterness and sourness.12 A shape may prime the matching taste and draw attention to it.12 In the online mug study, short mugs were expected to give more bitter and more intense coffee.13

Contrast. For the 2014 mug study, the authors suggest that a white mug may change how brown the coffee looks. That colour contrast may have changed perceived sweetness and intensity.8 They did not test this directly.

Odour–taste learning. In the 1999 model study, the effect of an odour on taste was best predicted by how sweet the odour smelled.6 Caramel odour also reduced the sourness of citric acid.6 The authors linked the results to previously learned associations between odours and tastes.6 A related effect, oral referral, makes odours seem to come from the mouth.5

The cup-texture study in detail#

The 2020 study ran three tests.10 In a pre-test, 31 coffee experts tasted while rubbing sandpaper with the other hand; aftertaste was rated more intense.10 The two conditions were not counterbalanced.10 In the main experiments, 81 Q graders and 117 amateurs each drank one specialty coffee from either a smooth or a rough cup. Ratings used 10 cm line scales from 0 to 10.10

0246810mean rating, 0–10 line scaleAftertaste dryness, Qgraders2.96.610 Aftertaste dryness,amateurs4.3 · 5.310 Acidity, Q graders5.2 · 6.010 Acidity, amateurs (n.s.)5.3 · 5.610 Sweetness, Q graders(n.s.)6.4 · 6.310 Sweetness, amateurs6.1 · 5.210 Body, Q graders (n.s.)4.3 · 5.010 Body, amateurs (n.s.)3.3 · 3.510
Data table
Pointmean rating, 0–10 line scaleSource
Aftertaste dryness, Q graders2.92.9
Aftertaste dryness, Q graders6.66.610
Aftertaste dryness, amateurs4.34.3
Aftertaste dryness, amateurs5.35.310
Acidity, Q graders5.25.2
Acidity, Q graders6.0610
Acidity, amateurs (n.s.)5.35.3
Acidity, amateurs (n.s.)5.65.610
Sweetness, Q graders (n.s.)6.46.4
Sweetness, Q graders (n.s.)6.36.310
Sweetness, amateurs6.16.1
Sweetness, amateurs5.25.210
Body, Q graders (n.s.)4.34.3
Body, Q graders (n.s.)5.0510
Body, amateurs (n.s.)3.33.3
Body, amateurs (n.s.)3.53.510
Mean ratings of one specialty coffee per group, drunk from a smooth porcelain cup (filled dot) or a rough ceramic cup (open dot), on 0–10 line scales. Aftertaste: 0 = smooth, 10 = dry. Paired values read smooth · rough. n.s. = not significant. Q graders and amateurs tasted different coffees. One study.10
Attribute (0–10) Q graders, smooth vs rough Amateurs, smooth vs rough
Aroma 6.5 vs 6.8, not significant10 5.8 vs 5.9, not significant10
Sweetness 6.4 vs 6.3, not significant10 6.1 vs 5.2, p < 0.0110
Acidity 5.2 vs 6.0, p < 0.0110 5.3 vs 5.6, not significant10
Body 4.3 vs 5.0, not significant10 3.3 vs 3.5, not significant10
Aftertaste (0 = smooth, 10 = dry) 2.9 vs 6.6, p < 0.000110 4.3 vs 5.3, p < 0.0110

Means as given in the paper’s table; the text gives some means to two decimals (6.67 for the rough-cup aftertaste of Q graders). The uncertainty the authors give for each mean is 0.2–0.3.10

For Q graders, the aftertaste effect was large: partial eta squared 0.63. The acidity effect was small, at 0.09.10 For amateurs, both the aftertaste and the sweetness effects had a partial eta squared of 0.07.10 Because each group tasted a different coffee, the authors note that experience and coffee type are confounded.10

Implications for cupping and research#

Vessels. SCA 102-2024 specifies cupping vessels of tempered glass or ceramic, 200–350 mL, with a top diameter of 75–90 mm.1 All cups must be identical in volume, dimensions and material.1 The standard does not specify colour or surface texture.1 Within one session, identical cups keep the vessel constant between samples; they do not remove its effect on the absolute scores.

Information. The CVA keeps physical results concealed from sensory assessors. Extrinsic information is compiled only after the sensory assessments are complete.2 A sensory textbook chapter adds coding, random or balanced order, repetitions and individual assessment.3

Comparing studies. Cup material, shape and colour are rarely the same between coffee studies. The cup studies above suggest that this can shift ratings.8,9,10 Serving temperature is a further variable: hotter coffee releases more volatile compounds.19

Open questions#

  • Replication. No study opened for this entry independently repeated a coffee cup effect with a new research team.
  • Size in practice. No study opened for this entry measured whether cup colour or texture changes scores on the SCA’s Coffee Value Assessment (CVA) forms.
  • Duration. The shape review notes that how long such effects last has not been studied.12
  • Coffee sweetness. As of June 2024, the SCA was researching whether coffee sweetness is a taste or a crossmodal effect of sweet-smelling compounds.2
  • Room and sound. Much of the evidence on lighting and music in coffee is unpublished or small.15

Key studies#

Van Doorn, Wuillemin and Spence, 2014 – Mug colour and café latte. In two small between-participants experiments, the same café latte was served in white, blue or transparent mugs. The white mug raised perceived intensity in one experiment and lowered perceived sweetness in the other; the authors suggest colour contrast as a cause.

Flavour 3, 10 · doi:10.1186/2044-7248-3-10

Carvalho and Spence, 2018 – Cup shape and specialty coffee. At a coffee event in Brazil, 276 professionals and amateurs tasted one coffee from a tulip, open or split cup. Aroma was strongest from the tulip cup; sweetness and acidity were highest from the split cup, which amateurs liked least.

Food Quality and Preference 68, 315–321 · doi:10.1016/j.foodqual.2018.04.003

Carvalho, Moksunova and Spence, 2020 – Cup texture. Q graders and amateurs drank specialty coffee from a smooth or a rough cup. Both groups rated the aftertaste drier from the rough cup; Q graders also rated it more acidic and amateurs less sweet, while body did not change.

Food Quality and Preference 81, 103841 · doi:10.1016/j.foodqual.2019.103841

Sörqvist et al., 2013 – An eco-label on identical coffee. Participants tasted two identical coffees, one described as “eco-friendly”. Those with strong sustainability attitudes preferred its taste and would pay more for it, also when they answered anonymously.

PLOS ONE 8(12), e80719 · doi:10.1371/journal.pone.0080719

Spence and Carvalho, 2019 – The setting of the coffee experience. A narrative review of how lighting, sound, scent and touch in the environment influence coffee taste and choice. It is open about the evidence base: some studies are unpublished or small.

Food Quality and Preference 80, 103802 · doi:10.1016/j.foodqual.2019.103802

Share or cite this entry

Change log

  • – Espresso study – "the sweeteners have no smell" replaced by "the sweeteners are not volatile". Reason: Verification; source 7 states that the sweeteners were nonvolatile, not that they were odourless.
  • – Unpublished music study reworded as suggested trends; "several" unpublished studies corrected to two (lighting, music). Reason: Verification; source 15 reports trends from seven participants and labels only these two studies as unpublished.
  • – Effects map – latte art arrow changed to dashed (expectations), cup shape to flavour intensity added as dashed; key moved to the caption only. Reason: Verification; the abstract of source 14 describes expectation and willingness-to-pay measures without tasting; source 13 reports expected intensity for short mugs.

Sources

  1. A/BSpecialty Coffee Association (2024). SCA Standard 102-2024 Coffee Value Assessment: Sample Preparation and Tasting Mechanics. Specialty Coffee Association.
  2. A/BSpecialty Coffee Association (writing team Fernández-Alduenda, M.; Ionescu, K. E.; Giuliano, P.; Rugolo, J.; Delrue, R.) (2024). A System to Assess Coffee Value: Understanding the Specialty Coffee Association’s Coffee Value Assessment (June 2024). Specialty Coffee Association.
  3. BThomas, E.; Puget, S.; Valentin, D.; Songer, P. (2017). Sensory Evaluation – Profiling and Preferences. In: Folmer, B. (ed.), The Craft and Science of Coffee, Academic Press, ch. 18, pp. 419–456. doi:10.1016/B978-0-12-803520-7.00018-9 · other copies
  4. ASpence, C. (2015). Multisensory flavor perception. Cell 161(1), 24–35.
  5. ASpence, C. (2016). Oral referral: On the mislocalization of odours to the mouth. Food Quality and Preference 50, 117–128.
  6. AStevenson, R. J.; Prescott, J.; Boakes, R. A. (1999). Confusing tastes and smells: how odours can influence the perception of sweet and sour tastes. Chemical Senses 24(6), 627–635.
  7. AChiralertpong, A.; Acree, T. E.; Barnard, J.; Siebert, K. J. (2008). Taste-Odor Integration in Espresso Coffee. Chemosensory Perception 1(2), 147–152.
  8. AVan Doorn, G. H.; Wuillemin, D.; Spence, C. (2014). Does the colour of the mug influence the taste of the coffee?. Flavour 3, 10. doi:10.1186/2044-7248-3-10 · other copies
  9. ACarvalho, F. M.; Spence, C. (2018). The shape of the cup influences aroma, taste, and hedonic judgements of specialty coffee. Food Quality and Preference 68, 315–321. doi:10.1016/j.foodqual.2018.04.003 · other copies
  10. ACarvalho, F. M.; Moksunova, V.; Spence, C. (2020). Cup texture influences taste and tactile judgments in the evaluation of specialty coffee. Food Quality and Preference 81, 103841. doi:10.1016/j.foodqual.2019.103841 · other copies
  11. ACarvalho, F. M.; Spence, C. (2019). Cup colour influences consumers’ expectations and experience on tasting specialty coffee. Food Quality and Preference 75, 157–169. doi:10.1016/j.foodqual.2019.03.001 · other copies
  12. ASpence, C.; Van Doorn, G. (2017). Does the Shape of the Drinking Receptacle Influence Taste/Flavour Perception? A Review. Beverages 3(3), 33. doi:10.3390/beverages3030033 · other copies
  13. AVan Doorn, G.; Woods, A.; Levitan, C. A.; Wan, X.; Velasco, C.; Bernal-Torres, C.; Spence, C. (2017). Does the shape of a cup influence coffee taste expectations? A cross-cultural, online study. Food Quality and Preference 56(A), 201–211. doi:10.1016/j.foodqual.2016.10.013 · other copies
  14. AVan Doorn, G.; Colonna-Dashwood, M.; Hudd-Baillie, R.; Spence, C. (2015). Latté art influences both the expected and rated value of milk-based coffee drinks. Journal of Sensory Studies 30(4), 305–315. doi:10.1111/joss.12159 · other copies
  15. ASpence, C.; Carvalho, F. M. (2019). The coffee drinking experience: product extrinsic (atmospheric) influences on taste and choice. Food Quality and Preference 80, 103802. doi:10.1016/j.foodqual.2019.103802 · other copies
  16. ASörqvist, P.; Hedblom, D.; Holmgren, M.; Haga, A.; Langeborg, L.; Nöstl, A.; Kågström, J. (2013). Who Needs Cream and Sugar When There Is Eco-Labeling? Taste and Willingness to Pay for “Eco-Friendly” Coffee. PLOS ONE 8(12), e80719. doi:10.1371/journal.pone.0080719 · other copies
  17. ABressani, A. P. P.; Martinez, S. J.; Batista, N. N.; Simão, J. B. P.; Schwan, R. F. (2021). Into the minds of coffee consumers: perception, preference, and impact of information in the sensory analysis of specialty coffee. Food Science and Technology (Campinas) 41(Suppl. 2), 667–675. doi:10.1590/fst.30720 · other copies
  18. ACarvalho, F. M.; de Sousa, M. M. M.; Nadaleti, D. H. S. (2026). Label Information About Fermentation Processing Affects Consumers’ Sensory and Hedonic Judgements of Specialty Coffee. Foods 15(8), 1287. doi:10.3390/foods15081287 · other copies
  19. ASteen, I.; Stolzenbach Wæhrens, S.; Agerlin Petersen, M.; Münchow, M.; Bredie, W. L. P. (2017). Influence of serving temperature on flavour perception and release of Bourbon Caturra coffee. Food Chemistry, 61–68.

Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.

Suggest a change

On Crossmodal effects. Tell us what you would change and why. A source that supports it helps most.