Abstract

Olfactory identification abilities in adolescents have been reported inferior compared with adults. Though this seems to be the case when comparing identification abilities using tests validated on—and for—adults, odor familiarity has been hypothesized to affect identification abilities in younger participants. However, this has never been thoroughly tested. The aims of this study were to investigate patterns in odor familiarity differences between adolescents and adults, and to investigate if an adolescent familiarity-based modification of an identification test could lead to similar identification scores in adolescents and adults. In total, 411 adolescent participants and 320 adult participants were included in the study. Odor familiarity ratings were obtained for 125 odors. A modified version of the “Sniffin’ Sticks” identification test was created and validated on 72 adolescents based on adolescent familiarity scores. This test was applied to 82 normosmic adults and 167 normosmic adolescents. Results show a lower familiarity for spices and environmental odors, and a higher familiarity for candy odors in adolescents. The identification abilities in adults and adolescents were equal after familiarity-based modification. We conclude that changes in odor familiarity from adolescence to adulthood do not develop evenly for all odors, but are dependent on odor-object category.

Introduction

Olfaction plays an inherent role in our perception of the world; however, its significance often is only acknowledged later in life. This is supported by the often delayed diagnosis of congenital anosmics (Qu et al. 2010) and by increased self-perceived significance of olfaction with increasing age (Oleszkiewicz et al. 2016b). When compared with adults, the olfactory scores of children are significantly lower, including the ability to identify odors (Hummel et al. 2006). This ability gradually increases during childhood and adolescence, reaching its peak in young adulthood (Cameron and Doty 2013; Sorokowska et al. 2015).

Normative data on olfactory identification abilities in adolescents are sparse. Most are based on olfactory tests that have been validated only in adult populations (Hummel et al. 2006; Hugh et al. 2015). As the odor identification abilities in children and adolescents are inferior to that of adults (Sorokowska et al. 2015), several studies have modified olfactory tests to be more applicable in younger populations, such as the modified 14-item “Sniffin’ Kids” identification test (Schriever et al. 2014), the candy test (Renner et al. 2009), and the game-like “ Smell wheel” odor identification test (Cameron and Doty 2013).

While these studies have made great efforts in adapting to this young target population, there is still much to be learned in order to understand how odor identification abilities mature during adolescence. The role of verbal abilities has previously been highlighted as a driving factor in the development of odor knowledge and perception in children (Monnery-Patris et al. 2009). A recent study on olfactory performance in adolescents also found an age-related co-development of odor identification abilities and verbal fluency (Oleszkiewicz et al. 2016b). The underlying mechanisms of how perceptual abilities affect the performance of adolescents in odor identification tests are still unknown. However, differences in odor familiarity have been mentioned as a possible confounder for lower olfactory identification scores in children and adolescents (Renner et al. 2009; Sorokowska et al. 2015). This could offer insights to the development of olfactory perception; however, this has never been thoroughly tested in adolescents.

To investigate the role of odor familiarity in odor identification differences between adolescents and adults, we conducted 3 substudies in order to test 2 hypotheses; firstly we hypothesize that there is a difference in odor familiarity between adolescents and adults, secondly, we hypothesize that by reducing nonfamiliar descriptors in an identification test, adolescents and adults will obtain comparable identification test scores. The aim of study 1 was to evaluate if there is a difference in odor familiarity of odor-object groups between adolescents and adults. The aims of study 2 were to modify an olfactory identification test in accordance with adolescent odor familiarity ratings, and validate the test in an adolescent population. The aim of study 3 was to evaluate if adolescents were still inferior to adults in identification test performance if the identification test had been modified in accordance with adolescent’s odor familiarity.

Materials and methods

Participants

Three steps were performed. A total of 731 participants were included in the 3 studies, of which 410 participated in the odor familiarity questionnaire study, and 321 underwent olfactory testing. The groups of participants were independent across studies. Characteristics for all participants are presented in Table 1.

Table 1.

Characteristics

Study 1. Odor familiarity study 
Group Age range n Gender (female/male) Mean odor familiarity 95% CI 
Adolescents 12–18 172 89/83 3.74 3.63–3.85 
 Younger 12–15 99 61/38 3.80 3.65–3.94 
 Older 16–18 73 28/45 3.67 3.49–3.84 
Adults 19–49 238 171/67 3.93 3.85–4.01 
 20’ 19–29 159 122/37 3.93 3.84–4.02 
 30’ 30–39 67 41/26 3.90 3.73–4.08 
 40’ 40–49 12 8/4 4.06 3.81–4.31 
Male gender 12–49 150 — 3.80 3.67–3.91 
 Adolescents 12–18 83 — 3.66 3.49–3.84 
 Adults 19–49 67 — 3.93 3.78–4.07 
Female gender 12–49 260 — 3.89 3.81–3.96 
 Adolescents 12–18 89 — 3.81 3.68–3.95 
 Adults 19–49 171 — 3.93 3.83–4.03 
Hyposmic 12–49 31 15/16 3.48 3.25–3.70 
 Adolescents 12–18 13 4/9 3.55 2.73–3.55 
 Adults 19–49 18 11/7 3.68 3.40–3.95 
Odor identification studies 
Group Age range (mean) n Gender (female/male) Mean identification score 95% CI 
Study 2. Sniffin’ Sticks evaluation 
 Evaluation group 12–18 (14.36) 72 41/31 13.46 13.06–13.86 
Study 3. Effect of familiarity and experience on identification scores 
 Adolescent 12–18 (14.88) 167 101/66 14.52 14.33–14.72 
 Adult 19–55 (32.37) 82 41/41 14.41 14.12–14.71 
Study 1. Odor familiarity study 
Group Age range n Gender (female/male) Mean odor familiarity 95% CI 
Adolescents 12–18 172 89/83 3.74 3.63–3.85 
 Younger 12–15 99 61/38 3.80 3.65–3.94 
 Older 16–18 73 28/45 3.67 3.49–3.84 
Adults 19–49 238 171/67 3.93 3.85–4.01 
 20’ 19–29 159 122/37 3.93 3.84–4.02 
 30’ 30–39 67 41/26 3.90 3.73–4.08 
 40’ 40–49 12 8/4 4.06 3.81–4.31 
Male gender 12–49 150 — 3.80 3.67–3.91 
 Adolescents 12–18 83 — 3.66 3.49–3.84 
 Adults 19–49 67 — 3.93 3.78–4.07 
Female gender 12–49 260 — 3.89 3.81–3.96 
 Adolescents 12–18 89 — 3.81 3.68–3.95 
 Adults 19–49 171 — 3.93 3.83–4.03 
Hyposmic 12–49 31 15/16 3.48 3.25–3.70 
 Adolescents 12–18 13 4/9 3.55 2.73–3.55 
 Adults 19–49 18 11/7 3.68 3.40–3.95 
Odor identification studies 
Group Age range (mean) n Gender (female/male) Mean identification score 95% CI 
Study 2. Sniffin’ Sticks evaluation 
 Evaluation group 12–18 (14.36) 72 41/31 13.46 13.06–13.86 
Study 3. Effect of familiarity and experience on identification scores 
 Adolescent 12–18 (14.88) 167 101/66 14.52 14.33–14.72 
 Adult 19–55 (32.37) 82 41/41 14.41 14.12–14.71 

Participants are listed for study 1–3 separately.

CI, confidence interval.

Study 1: analysis of developmental differences in odor familiarity

In this study, 410 (172 adolescents) participants rated how familiar they were with common odor descriptors in a questionnaire, along with age, gender, subjective assessment of olfactory capacity, and ethnicity. Age was registered in intervals (12–15; 16–18; 19–29; 30–39; 40–49). The questionnaire survey was conducted in 8 classes on 3 different elementary schools and a high school in different school districts in order to reduce the effects of socioeconomic status and related differences in food habits and odor familiarity on the results. The pupils had to rate their familiarity of 125 odor descriptors, including the original descriptors from the adult Sniffin’ Sticks test. Adult participants (n = 238) were recruited through QR codes in public places and online.

Study 2: evaluation of the adult Sniffin Sticks identification test in an adolescent population

In this study, 78 adolescents underwent olfactory evaluation with the Danish original Sniffin’ Stick 16-item identification test (SIT-16), where ethnicity, age, subjective olfactory capacity, allergies, and history of sinonasal disease were registered. The test scores of the evaluation group were used for modification of the Sniffin’ Stick 16-item identification test for adolescents (SIT-16jr) in combination with adolescent familiarity scores.

Study 3: the effect of familiarity and experience on odor identification scores

In this study, 86 adults (age 19–55) and 173 adolescents (age 12–18) underwent olfactory evaluation with the SIT-16jr, where the following parameters were registered: ethnicity, age, subjective olfactory capacity, allergies, and history of sinonasal disease. Only normosmic participants were included in study for further analysis (Hummel et al. 1997), resulting in a total of 249 participants.

Ethical considerations

The olfactory identification test studies were approved by the Danish Ethical Committee (Etisk komité, Central Denmark Region). Ethics approval was given for olfactory testing (1-10-72-337-13). No approval from an ethical committee was required for the questionnaire according to Danish law (Komitéloven § 14, stk. 2.). Both studies were conducted in accordance with the Declaration of Helsinki ethical principles for medical research. Informed consent for participating in the olfactory identification test was given prior to enrolment by the participant, and for the participants under the legal age of 18 years, parents (custody holders) also gave informed consent.

Procedures

Study 1: analysis of developmental differences in odor familiarity

Participants used a Likert-type scale to rate odor familiarity for 125 common odor descriptors. The Likert-type scale ranged from 1–5, with the following instruction to participants: On a scale from 1–5, please rate how familiar you are with the odor. 1) I would not be able to recognize the odor; 2) I do not think I would be able to recognize the odor; 3) Maybe I would be able to recognize the odor; 4) I think I would be able to recognize the odor; 5) I would be able to recognize the odor. The odorants included in the questionnaire were descriptors from the Sniffin’ Sticks extended identification test, combined with other common food-related and environmental odor descriptors. The 125 odors were prospectively placed in 10 “odor object categories”: environmental, candy, sweet foods, nuts, acrid foods, meat/fish, vegetables, alcohol, other foods, and spices/seasoning. The definition of categories and initial placement of odor in each category was made in collaboration with 2 chefs, followed by a validation of the placement of all odors in odor-object categories in a focus group of 12 adolescents (5 male). The odors of the 10 odor-object categories were equally distributed throughout the questionnaire to avoid differences in odor-object category scores due to attention bias. Data collection was made through a professional online questionnaire service (Survey-Xact.dk, Ramboll Management Consulting A/S, Denmark).

Study 2: adaptation and validation of Sniffin Sticks in an adolescent population

The SIT-16 contains 16 odorants in felt-tip-like devices. The test procedure for each odorant was initiated by informing the participant about the test procedure, including that they were allowed to sample each odorant twice before choosing a descriptor. Then the participant were instructed to read the 4 possible descriptors, subsequently, the experimenter removed the cap and placed the tip of the felt-tip pen 1–2 cm in front of both nostrils of the participant for 3–4 s. The participant then had to select the correct odorant between the 4 possible descriptors in a forced multiple-choice test. Each descriptor choice was registered and an identification score ranging from 0 to 16 was calculated (Kobal et al. 1996).

The modification process had 2 steps. Firstly, it was evaluated if the correct identification rate for all 16 odorants in a normosmic adolescent population reached the required 75% (Hummel et al. 1997). Secondly, descriptors were modified from 2 criteria; all unfamiliar descriptors were altered to odor descriptors with a ≥75% familiarity rate. Additionally, in the modification of descriptors for odorants with a low familiarity rating, emphasis was put on pairing the less familiar odorant with highly familiar odor descriptors with a higher degree of contrast between descriptors (Fjaeldstad et al. 2015).

Study 3: effects of adolescent Sniffin’ Sticks modification on adult identification rates

The SIT16jr was applied to an adult and an adolescent population after modification of odorant descriptors based on adolescent’s familiarity scores. The application procedures were identical to the described procedures in study 2.

Statistics

Results were analyzed with JMP 13 for Mac (SAS Institute Inc., NC, USA). For multiple comparisons of the prospectively defined primary aim of study 1, the alpha levels underwent Bonferroni adjustment (Moyé 2008). For multiple comparison testing, the P values deemed significant were presented in conjunction with relevant statistics. For secondary endpoints and post hoc analysis, P values of <0.05 were considered statistically significant. Normal distribution was assessed using QQ-plots and Shapiro–Wilk W test. In parametric data, mean values were compared using the student’s t-test, while nonparametric data were compared using the Cochran test, Wilcoxon rank-sum test, and Kruskal–Wallis test as appropriate. Fisher’s Exact Test (FET) was used for contingency analysis. Degrees of freedom are written in subscript when indicated.

Results

Study 1: analysis of developmental differences in odor familiarity

In this study, participants rated how familiar they were with 125 common odors in a questionnaire (Table 2). Each odor was included in 1 of the 10 predefined odor object categories, which in combination with age group were used to analyze familiarity differences between adolescents and adults (Table 3; Supplementary table).

Table 2.

Adolescent and adult odor descriptor familiarity ratings

Odor %/% Odor %/% Odor %/% Odor  %/% 
Petrol 98/95 33 Marzipan 78/81 65 Bread 62/87 97 Mushroom 46/50 
Coffee 95/98 34 Felt-tip pen 78/89 66 Butter 62/47 98 Green capsicum 46/42 
Curry 93/92 35 Peppermint 78/87 67 Fruit gum 61/51 99 Dill 45/80 
Fish 92/94 36 Grass 77/94 68 Smoked meat 61/61 100 Peach 45/54 
Cinnamon 92/99 37 Strawberry 77/83 69 Ham 61/55 101 Blackcurrant 45/42 
Popcorn 90/92 38 Pineapple 76/92 70 Parsley 61/71 102 Peas 45/44 
Cigarette 90/97 39 Caramel 75/69 71 Honeydew melon 61/54 103 Almond 43/46 
Paint 90/86 40 Elder flower 75/79 72 Grapes 61/40 104 Grapefruit 43/54 
Bacon 90/97 41 Stable 74/82 73 Carrot 61/51 105 Hazelnut 41/37 
10 Cheese 90/96 42 Paprika 73/56 74 Whisky 60/53 106 Broccoli 41/37 
11 Smoke 89/94 43 Smoked salmon 72/78 75 Hay 60/74 107 Avocado 39/43 
12 Vanilla 89/95 44 Rye bread 72/77 76 Raspberry 60/64 108 Blackberry 37/27 
13 Orange 88/99 45 Coconut 72/84 77 Corn 60/44 109 Plum 36/30 
14 Garlic 87/97 46 Cookie 72/63 78 Tomato 60/61 110 Cranberry 36/28 
15 Onion 87/85 47 Apple 72/75 79 Marshmallow 58/32 111 Clove 34/58 
16 Sweat 86/92 48 Salami 71/69 80 Rose 58/69 112 Thyme 33/67 
17 Burnt almond 86/89 49 Wood 70/86 81 Camomile 58/75 113 Lilac 32/38 
18 Steak 85/84 50 Ginger 70/85 82 Chives 57/71 114 Balsamic vinegar 32/68 
19 Wine 84/88 51 Banana 70/86 83 Cucumber 57/77 115 Walnut 31/42 
20 Spruce 84/94 52 Water melon 68/49 84 Capsicum 57/54 116 Eucalyptus 30/74 
21 Soap 84/87 53 Glue 67/70 85 Snickers bar 56/34 117 Lavender 29/74 
22 Urine 82/91 54 Oregano 66/82 86 Basil 56/85 118 Celery 28/50 
23 Ketchup 82/72 55 Mustard 66/72 87 Mars bar 54/31 119 Nutmeg 26/54 
24 Coke 81/72 56 Beetroot 66/46 88 Nougat 54/43 120 Asparagus 26/46 
25 Wet dog 81/81 57 Soy 65/64 89 Pear 54/58 121 Turpentine 24/53 
26 Mint 81/84 58 Melon 65/50 90 Raisin 53/56 122 Sauerkraut 21/17 
27 Pepper 81/82 59 Candyfloss 64/39 91 Mango 52/61 123 Tarragon 20/50 
28 Pancake 81/74 60 Rubber 64/68 92 Cherry 51/51 124 Truffle 20/24 
29 Chocolate 80/78 61 Leather 64/85 93 Seaweed 48/53 125 Anise 14/65 
30 Honey 80/74 62 Rum 63/60 94 Whipped cream 48/46    
31 Lemon 79/92 63 Peanuts 63/66 95 Pickles 48/49    
32 Liquorice 78/83 64 Menthol 63/83 96 Cabbage 47/46    
Odor %/% Odor %/% Odor %/% Odor  %/% 
Petrol 98/95 33 Marzipan 78/81 65 Bread 62/87 97 Mushroom 46/50 
Coffee 95/98 34 Felt-tip pen 78/89 66 Butter 62/47 98 Green capsicum 46/42 
Curry 93/92 35 Peppermint 78/87 67 Fruit gum 61/51 99 Dill 45/80 
Fish 92/94 36 Grass 77/94 68 Smoked meat 61/61 100 Peach 45/54 
Cinnamon 92/99 37 Strawberry 77/83 69 Ham 61/55 101 Blackcurrant 45/42 
Popcorn 90/92 38 Pineapple 76/92 70 Parsley 61/71 102 Peas 45/44 
Cigarette 90/97 39 Caramel 75/69 71 Honeydew melon 61/54 103 Almond 43/46 
Paint 90/86 40 Elder flower 75/79 72 Grapes 61/40 104 Grapefruit 43/54 
Bacon 90/97 41 Stable 74/82 73 Carrot 61/51 105 Hazelnut 41/37 
10 Cheese 90/96 42 Paprika 73/56 74 Whisky 60/53 106 Broccoli 41/37 
11 Smoke 89/94 43 Smoked salmon 72/78 75 Hay 60/74 107 Avocado 39/43 
12 Vanilla 89/95 44 Rye bread 72/77 76 Raspberry 60/64 108 Blackberry 37/27 
13 Orange 88/99 45 Coconut 72/84 77 Corn 60/44 109 Plum 36/30 
14 Garlic 87/97 46 Cookie 72/63 78 Tomato 60/61 110 Cranberry 36/28 
15 Onion 87/85 47 Apple 72/75 79 Marshmallow 58/32 111 Clove 34/58 
16 Sweat 86/92 48 Salami 71/69 80 Rose 58/69 112 Thyme 33/67 
17 Burnt almond 86/89 49 Wood 70/86 81 Camomile 58/75 113 Lilac 32/38 
18 Steak 85/84 50 Ginger 70/85 82 Chives 57/71 114 Balsamic vinegar 32/68 
19 Wine 84/88 51 Banana 70/86 83 Cucumber 57/77 115 Walnut 31/42 
20 Spruce 84/94 52 Water melon 68/49 84 Capsicum 57/54 116 Eucalyptus 30/74 
21 Soap 84/87 53 Glue 67/70 85 Snickers bar 56/34 117 Lavender 29/74 
22 Urine 82/91 54 Oregano 66/82 86 Basil 56/85 118 Celery 28/50 
23 Ketchup 82/72 55 Mustard 66/72 87 Mars bar 54/31 119 Nutmeg 26/54 
24 Coke 81/72 56 Beetroot 66/46 88 Nougat 54/43 120 Asparagus 26/46 
25 Wet dog 81/81 57 Soy 65/64 89 Pear 54/58 121 Turpentine 24/53 
26 Mint 81/84 58 Melon 65/50 90 Raisin 53/56 122 Sauerkraut 21/17 
27 Pepper 81/82 59 Candyfloss 64/39 91 Mango 52/61 123 Tarragon 20/50 
28 Pancake 81/74 60 Rubber 64/68 92 Cherry 51/51 124 Truffle 20/24 
29 Chocolate 80/78 61 Leather 64/85 93 Seaweed 48/53 125 Anise 14/65 
30 Honey 80/74 62 Rum 63/60 94 Whipped cream 48/46    
31 Lemon 79/92 63 Peanuts 63/66 95 Pickles 48/49    
32 Liquorice 78/83 64 Menthol 63/83 96 Cabbage 47/46    

Descriptors are regarded as familiar if the participant rated the odor “familiar” or “highly familiar” on the Likert scale. %/% = Adolescent familiarity in percent/Adult familiarity in percent.

Table 3.

Age-related familiarity differences of between odor object categories

Odor object category Amount of odors Mean familiarity score (95% CI) P value 
(nAdult Adolescent Difference 
Food-related odors 
 Candy 12 3.74 3.96 −0.23 (−0.38 to −0.07) 0.0050 
 Sweet foods 24 3.73 3.72 0.01 (−0.15 to − 0.17) 0.8615 
 Nuts 3.68 3.49 0.19 (0.02 to 0.37) 0.0321 
 Meat/fish 4.22 4.17 0.06 (−0.08 to 0.20) 0.4139 
 Acrid foods 4.26 3.84 0.43 (0.29 to 0.56) <0.0001 
 Vegetables 17 3.60 3.48 0.12 (−0.04 to 0.28) 0.1507 
 Spices/seasoning 20 4.05 3.48 0.57 (0.43 to 0.71) <0.0001 
 Other foods 3.67 3.57 0.10 (−0.07 to 0.26) 0.2475 
Nonfood odors 
 Alcohol 3.90 3.88 0.02 (−0.18 to 0.21) 0.8543 
 Environmental 23 4.29 3.98 0.31 (0.19 to 0.43) <0.0001 
Odor object category Amount of odors Mean familiarity score (95% CI) P value 
(nAdult Adolescent Difference 
Food-related odors 
 Candy 12 3.74 3.96 −0.23 (−0.38 to −0.07) 0.0050 
 Sweet foods 24 3.73 3.72 0.01 (−0.15 to − 0.17) 0.8615 
 Nuts 3.68 3.49 0.19 (0.02 to 0.37) 0.0321 
 Meat/fish 4.22 4.17 0.06 (−0.08 to 0.20) 0.4139 
 Acrid foods 4.26 3.84 0.43 (0.29 to 0.56) <0.0001 
 Vegetables 17 3.60 3.48 0.12 (−0.04 to 0.28) 0.1507 
 Spices/seasoning 20 4.05 3.48 0.57 (0.43 to 0.71) <0.0001 
 Other foods 3.67 3.57 0.10 (−0.07 to 0.26) 0.2475 
Nonfood odors 
 Alcohol 3.90 3.88 0.02 (−0.18 to 0.21) 0.8543 
 Environmental 23 4.29 3.98 0.31 (0.19 to 0.43) <0.0001 

Conservative measures with 2-tailed t-test were used, t408 (Mean Odor-object score was calculated for each participant). CI, confidence interval.

P values of ≤0.005 were regarded as significant after Bonferroni adjustment.

Age had a significant effect on rating odor familiarity. Adolescents (n = 172) rated the odor familiarity with a mean Likert-score of 3.74 (SD = 0.74), while adults (n = 238) had a mean Likert-score of 3.93 (SD = 0.62), (t408 = 0.19, P = 0.0051, dCohen = 0.282). Segmenting familiarity rating into the predefined odor object categories, allowed for comparison of perception of certain types of odors between adolescents and adults. Even though conservative measures were taken for the multiple comparison analysis of odor object categories, we found significant age-related differences in odors related to spices and seasoning as well as in environmental odors. Adolescents were more familiar with candy odors, although this was no longer regarded as significant after Bonferroni adjustment. In 3 odor-object categories, we found a null mean-familiarity difference, while the remaining 4 categories showed variation between adolescents and adults (Table 3).

A distribution plot of mean familiarity between odor object categories revealed a reoccurring low-mean tail effect in the categories where adolescents are lacking behind the adult familiarity scores. This is observed for vegetables, acrid foods, environmental odors, and spices/seasoning (Figure 1).

Figure 1.

Odor familiarity ratings in adolescents and adults. Graphs of the relations between odor familiarity in adolescents and adults for 6 odor object groups. Please notice the steep slope of the curve for candy and sweet food odors and the flat slope of the curve for spices and environmental odors, indicating a large difference in familiarity between odor-object groups.

Figure 1.

Odor familiarity ratings in adolescents and adults. Graphs of the relations between odor familiarity in adolescents and adults for 6 odor object groups. Please notice the steep slope of the curve for candy and sweet food odors and the flat slope of the curve for spices and environmental odors, indicating a large difference in familiarity between odor-object groups.

There were no significant differences in odor familiarity rating between genders (t408 = 0.11, P = 0.26, dCohen = 0.164) or ethnicity (t408 = 0.19, P = 0.17, dCohen = 0.039). Participants who described themselves as subjective hyposmic (n = 31) rated the odors as significantly less familiar compared with the subjectively normosmic participants (t408 = −0.43, P = 0.0005, dCohen = 0.652). The subjectively hyposmic adolescents and adult participants had an equally distributed lower mean score for both environmental and food-related odors. The only odorant descriptors hyposmic individuals rated equally high as normosmic individuals, were gasoline, cigarettes, fish, garlic, cinnamon, paint, and sweat. No odor descriptors were significantly lower in self-reported hyposmic individuals.

Study 2: evaluation and modification of Sniffin’ Sticks in an adolescent population

In the normosmic adolescent evaluation group (n = 72), the mean Sniffin’ Sticks identification test (SIT) score was 13.76 (SD = 1.50); with a male average of 13.77 (SD = 1.48), a female mean score of 13.76 (SD = 1.53), and no significant difference between male (n = 31) and female (n = 41) (z70 = 0.076, P = 0.94, η2 = 0.0001). Hyposmic individuals were excluded prior to analysis (6 participants, 4 male).

There were significant differences between the identification rates of the 16 odors (Q15 = 138.68, P < 0.0001). Three odors were systematically identified as one of the false descriptors causing the identification rate to drop for the following odorants: coffee (74% correct), pineapple (67% correct), and apple (56% correct). The coffee odorant was systematically identified as the incorrect cigarette descriptor, and though coffee and cigarette odors are highly familiar to adolescents (Table 2), the cigarette odor familiarity is significantly lower for adolescents than for adults (t408 = 0.25, P = 0.0009, d = 0.025).

The pineapple odor was rated “familiar” or “highly familiar” by 72% of adolescents, which was significantly lower than the adult familiarity (t408 = 0.49, P = < 0.0001). In combination with the option of pear as another sweet descriptor choice, which also had low familiarity in adolescents, only 67% of adolescents were able to correctly identify the pineapple odor.

The apple odor was also rated “familiar” or “highly familiar” by 72% of adolescents in the familiarity questionnaire; however, only 56% were able to correctly identify the apple descriptor in the evaluation group, where 36% falsely identified the odorant as strawberry.

From the odor familiarity study, we learned that several of the descriptors were unfamiliar to the adolescent population and the non-odorant descriptors were modified prior to study 3. Furthermore, 7 of the Sniffin’ Sticks odorants were unfamiliar in adolescents. Among all descriptors, only 68.8% were familiar for adolescents and 83.5% for adults. The correct descriptors of Sniffin’ Sticks with unfamiliar odorants could not be modified. However, after modifying the test, the average familiarity score among the other odor descriptors were 82.9% for adolescents and 84.5% for adults.

Study 3: effects of adolescent Sniffin’ Sticks modification on adult identification rates

As normosmia was a requirement for participation (Identification score ≥11 (Hummel et al. 1997), equivalent to the 10th percentile in our data), 6 adolescents and 4 adults were excluded from the study. The mean identification score in the adult population (n = 82) was 14.41 (SD = 1.35), while the adolescent population (n = 167) had a mean identification score of 14.52 (SD = 1.28). Although adolescents had a higher mean identification score, this was not statistically significant (z247 = 0.54, P = 0.5861, η2 = 0.001). The only odorant where the identification rate was significantly dependent on age group was cinnamon (FET, P = 0.0022), where adolescents had a higher identification score than adults, as adolescents did not confuse the cinnamon odorant for the vanilla descriptor (3.6% error vs. 12.2% error in adults) or chocolate descriptor (2.4% error vs. 8.5% error in adults) (Figure 2).

Figure 2.

Odorant identification rates for the SIT-16jr in adults (red) and adolescents (blue). After modification based on adolescent familiarity ratings, there was a similar identification score for adolescents and adults. However, adolescents were still significantly better than adults at discriminating cinnamon from the distractors (vanilla and chocolate)—odors commonly found in candy.

Figure 2.

Odorant identification rates for the SIT-16jr in adults (red) and adolescents (blue). After modification based on adolescent familiarity ratings, there was a similar identification score for adolescents and adults. However, adolescents were still significantly better than adults at discriminating cinnamon from the distractors (vanilla and chocolate)—odors commonly found in candy.

Female adolescents (n = 101) had slightly higher identification score (M = 14.65, SD = 1.18) than male adolescents (n = 66) (M = 14.32, SD = 1.41), although this did not reach statically significance (z165 = 1.40, P = 0.1603, η2 = 0.012).

Discussion

The main finding of this study was the existence of odor-object dependent changes in odor familiarity between adolescents and adults; adults had higher odor familiarity scores in spice- and environmental-related odors, while adolescents had a higher odor familiarity score for candy. When correcting for these findings in a modified odor identification test, adolescents had odor identification test scores comparable with adults.

Odor familiarity

By investigating the differences in odor familiarity between adolescents and adults, we identified significant differences in familiarity for certain odor object groups. Adolescents had significantly lower mean familiarity scores for spices/seasoning, environmental odors, and acrid foods, but a higher mean familiarity score for candy (Table 3). The distribution of the lower familiarity scores in the vegetable and acrid foods odor object groups revealed a tail pattern, where odors that were slightly less familiar in adults were highly unfamiliar in adolescents (Figure 1).

Subjectively hyposmic individuals generally rated descriptors as less familiar, when compared with normosmic individuals. However, this did not apply for gasoline, cigarettes, fish, garlic, cinnamon, paint, and sweat, suggesting an important role of trigeminal sensation when interpreting differences in olfactory abilities (Frasnelli et al. 2007). From the perspective of differentiating normosmics from hyposmics, we found that no odor descriptors were superior in discriminating self-reported hyposmia from normosmia.

Effects of familiarity in odor identification tests

The odor familiarity scores from study 1 were applied in the modification process of the SIT-16. Although some odorants in the SIT-16 were not familiar to adolescents, the data from the first 2 studies were used to create an olfactory identification test modified for adolescents. The unfamiliar false distractors were replaced, and unfamiliar odorant distractors were paired with highly familiar false distractors. Thus, while prior studies have compared identification scores of adolescents with those of adults by applying a test specifically validated for adults (Hummel et al. 2006; Schriever et al. 2014; Sorokowska et al. 2015), we have turned the tables, and tested the 2 age groups with a test validated for adolescents. We found that in this setting, adolescents were not inferior to adults. Furthermore, the only odorant with a significant difference between age groups was cinnamon, where the adults mistook the odorant for the sweet-related scents of vanilla and chocolate—an error the adolescents were not prone to make.

As odor identification may be influenced by odor familiarity, this could have an effect in the olfactory tests applied in children and adolescents, where there is a high frequency of flower, environmental, and spice/herbal odorants and descriptors [see e.g., (Monnery-Patris et al. 2009) or (Schriever et al. 2014)]. Unlike our findings, where adolescents were not inferior to adults in odor identification, the presence of highly unfamiliar odorants and distractors are likely to cause an effect modification by age, as suggested previously in the literature (Sorokowska et al. 2015). Our findings in adolescents are in line with the hypothesis that children may lack odor-specific knowledge which accumulates throughout life (Monnery-Patris et al. 2009).

Olfactory learning

We are only beginning to understand the link from perception through learning to correct identification of an odor. Though odors are potent triggers of autobiographical memories from as far back as the first decade of life (Arshamian et al. 2013)—and in a way closely linked to memory—the ability to name odors is an acquired skill that takes years to master and requires both perceptual and verbal training (Castriota-Scanderbeg et al. 2005; Plailly et al. 2012). For experts within a certain field, such as baristas or sommeliers, training and repeated exposure is key for improving this skill, however, they selectively become better at the smells they practise (Croijmans and Majid 2016). This is in line with a previous study, where perceptual experience modified cortical odor perception (Li et al. 2006). From this perspective, the age-related increasing significance of olfaction in adolescents (Oleszkiewicz et al. 2016b) may be correlated with our findings of age-related improvements of odor familiarity.

It may be hypothesized that the domain-specific findings in people who have undergone specific perceptual olfactory training (such as wine recognition in wine experts or chemical compound recognition in perfumers), may be more universal and that improvement of olfactory identification skills may simply mirror cumulative exposure of odorants. Studies of patients with olfactory disorders reveal sustainable improvements of odor identification after olfactory training (Konstantinidis et al. 2016). The effects of training may be a result of plasticity in the olfactory bulb, which has been shown to have a high degree of neural plasticity (Huart et al. 2013), or it may be caused by more central alterations of olfactory processing (Kollndorfer et al. 2015). Though we cannot deduct the underlying neural mechanisms of development of odor familiarity, we conclude that the development of odor familiarity from adolescence to adulthood does not occur equally for all odors, but is dependent on odor-object category.

Implications for future research

Firstly, our findings call for considering odor familiarity when modifying an odor identification test for adolescents. Normally, before olfactory identification tests are implemented in a new language or cultural setting, they undergo modification and validation to ensure that the test is valid in the new target population. Thus, it is common practice to formally evaluate culturally based differences in odor familiarity and change descriptors accordingly (Yuan et al. 2010; Oleszkiewicz et al. 2016a). However, when changing the target population to adolescents, previous studies have often not adjusted for the familiarity of descriptors (Monnery-Patris et al. 2009; Schriever et al. 2014; Sorokowska et al. 2015). Secondly, odor familiarity seems to play a key role in the development of odor identification abilities. In the further studies of learning in olfaction, familiarity could be included as a covariate in order to gain further insight to the mechanisms and development of olfactory processing.

To our knowledge, this is the first study to investigate differences in odor familiarity between adolescence and adulthood. This may add a small piece to the puzzle in understanding the development of olfactory abilities in adolescents.

Supplementary material

Supplementary data are available at Chemical Senses online.

Author contributions

Conceived and designed the experiments: A.F. (study 1–3), A.F. and T.O. (study 2). Performed the experiments: A.F., J.S., A.N., and T.O. Analyzed the data: A.F. Wrote first draft of the paper: A.F. Critically revising of the paper: A.F., J.S., A.N., T.O.

Conflict of interest

The authors declare to have no conflict of interest. There was no external funding for the project.

Acknowledgements

We are grateful to the children, their parents, and to all participants for their contribution. Furthermore, we would like to acknowledge Skaade School, Frijsenborg boarding school, and Eriksminde boarding school, Dorthe Rasmussen, Arne Moeller, and Monika Holm Nissen for their help with organizing the olfactory testing.

References

Arshamian
A
Iannilli
E
Gerber
JC
Willander
J
Persson
J
Seo
HS
Hummel
T
Larsson
M
.
2013
.
The functional neuroanatomy of odor evoked autobiographical memories cued by odors and words
.
Neuropsychologia
 .
51
(
1
):
123
131
.
Cameron
EL
Doty
RL
.
2013
.
Odor identification testing in children and young adults using the smell wheel
.
Int J Pediatric Otorhinolaryngol
 .
77
:
346
350
.
Castriota-Scanderbeg
A
Hagberg
GE
Cerasa
A
Committeri
G
Galati
G
Patria
F
Pitzalis
S
Caltagirone
C
Frackowiak
R
.
2005
.
The appreciation of wine by sommeliers: a functional magnetic resonance study of sensory integration
.
Neuroimage
 .
25
(
2
):
570
578
.
Croijmans
I
Majid
A
.
2016
.
Not all flavor expertise is equal: the language of wine and coffee experts
.
PLoS One
 .
11
(
6
):
e0155845
.
Fjaeldstad
A
Kjaergaard
T
Van Hartevelt
TJ
Moeller
A
Kringelbach
ML
Ovesen
T
.
2015
.
Olfactory screening: validation of Sniffin’ Sticks in Denmark
.
Clin Otolaryngol
 .
40
(
6
):
545
550
.
Frasnelli
J
Schuster
B
Hummel
T
.
2007
.
Interactions between olfaction and the trigeminal system: what can be learned from olfactory loss
.
Cereb Cortex
 .
17
(
10
):
2268
2275
.
Huart
C
Rombaux
P
Hummel
T
.
2013
.
Plasticity of the human olfactory system: the olfactory bulb
.
Molecules
 .
18
(
9
):
11586
11600
.
Hugh
SC
Siu
J
Hummel
T
Forte
V
Campisi
P
Papsin
BC
Propst
EJ
.
2015
.
Olfactory testing in children using objective tools: comparison of Sniffin’ Sticks and University of Pennsylvania Smell Identification Test (UPSIT)
.
J Otolaryngol Head Neck Surg
 .
44
:
10
.
Hummel
T
Kobal
G
Gudziol
H
Mackay-Sim
A
.
2007
.
Normative data for the “Sniffin’ Sticks” including tests of odor identification, odor discrimination, and olfactory thresholds: an upgrade based on a group of more than 3,000 subjects
.
Eur Arch Otorhinolaryngol
 .
264
(
3
):
237
243
.
Hummel
T
Sekinger
B
Wolf
SR
Pauli
E
Kobal
G
.
1997
.
‘Sniffin’ sticks’: olfactory performance assessed by the combined testing of odor identification, odor discrimination and olfactory threshold
.
Chem Senses
 .
22
(
1
):
39
52
.
Kobal
G
Hummel
T
Sekinger
B
Barz
S
Roscher
S
Wolf
S
.
1996
.
“Sniffin’ sticks”: screening of olfactory performance
.
Rhinology
 .
34
(
4
):
222
226
.
Kollndorfer
K
Fischmeister
FP
Kowalczyk
K
Hoche
E
Mueller
CA
Trattnig
S
Schöpf
V
.
2015
.
Olfactory training induces changes in regional functional connectivity in patients with long-term smell loss
.
Neuroimage Clin
 .
9
:
401
410
.
Konstantinidis
I
Tsakiropoulou
E
Constantinidis
J
.
2016
.
Long term effects of olfactory training in patients with post-infectious olfactory loss
.
Rhinology
 .
54
(
2
):
170
175
.
Li
W
Luxenberg
E
Parrish
T
Gottfried
JA
.
2006
.
Learning to smell the roses: experience-dependent neural plasticity in human piriform and orbitofrontal cortices
.
Neuron
 .
52
(
6
):
1097
1108
.
Monnery-Patris
S
Rouby
C
Nicklaus
S
Issanchou
S
.
2009
.
Development of olfactory ability in children: sensitivity and identification
.
Dev Psychobiol
 .
51
(
3
):
268
276
.
Moyé
LA
.
2008
.
Disciplined analyses in clinical trials: the dark heart of the matter
.
Stat Methods Med Res
 .
17
:
253
264
.
Oleszkiewicz
A
Taut
M
Sorokowska
A
Radwan
A
Kamel
R
Hummel
T
.
2016
a.
Development of the Arabic version of the “Sniffin’ Sticks” odor identification test
.
Eur Arch Otorhinolaryngol
 .
273
(
5
):
1179
1184
.
Oleszkiewicz
A
Walliczek-Dworschak
U
Klötze
P
Gerber
F
Croy
I
Hummel
T
.
2016
b.
Developmental changes in adolescents’ olfactory performance and significance of olfaction
.
PLoS One
 .
11
(
6
):
e0157560
.
Plailly
J
Delon-Martin
C
Royet
JP
.
2012
.
Experience induces functional reorganization in brain regions involved in odor imagery in perfumers
.
Hum Brain Mapp
 .
33
(
1
):
224
234
.
Qu
Q
Liu
J
Ni
D
Zhang
Q
Yang
D
Wang
N
Wu
X
Han
H
.
2010
.
Diagnosis and clinical characteristics of congenital anosmia: case series report
.
J Otolaryngol Head Neck Surg
 .
39
(
6
):
723
731
.
Renner
B
Mueller
CA
Dreier
J
Faulhaber
S
Rascher
W
Kobal
G
.
2009
.
The candy smell test: a new test for retronasal olfactory performance
.
Laryngoscope
 .
119
(
3
):
487
495
.
Schriever
VA
Mori
E
Petters
W
Boerner
C
Smitka
M
Hummel
T
.
2014
.
The “Sniffin’ Kids” test–a 14-item odor identification test for children
.
PLoS One
 .
9
(
6
):
e101086
.
Sorokowska
A
Schriever
VA
Gudziol
V
Hummel
C
Hähner
A
Iannilli
E
Sinding
C
Aziz
M
Seo
HS
Negoias
S
et al
.
2015
.
Changes of olfactory abilities in relation to age: odor identification in more than 1400 people aged 4 to 80 years
.
Eur Arch Otorhinolaryngol
 .
272
(
8
):
1937
1944
.
Yuan
B-C
Lee
P-L
Lee
Y-L
Lin
S-H
Shu
C-H
.
2010
.
Investigation of the Sniffin’ Sticks olfactory test in Taiwan and comparison with different continents
.
J Chin Med Assoc
 .
73
:
483
486
.

Author notes

Correspondence to be sent to: Alexander Fjaeldstad, Aarhus University, Nørrebrogade 44, Bldg 10G, 2nd floor, 8000 Aarhus C, Denmark. e-mail: alexander.fjaeldstad@psych.ox.ac.uk

Supplementary data