An olfactory demography of a diverse metropolitan population

Abstract

Background: Human perception of the odour environment is highly variable. People vary both in their general olfactory acuity as well as in if and how they perceive specific odours. In recent years, it has been shown that genetic differences contribute to variability in both general olfactory acuity and the perception of specific odours. Odour perception also depends on other factors such as age and gender. Here we investigate the influence of these factors on both general olfactory acuity and on the perception of 66 structurally and perceptually different odours in a diverse subject population.

Results: We carried out a large human olfactory psychophysics study of 391 adult subjects in metropolitan New York City, an ethnically and culturally diverse North American metropolis. 210 of the subjects were women and the median age was 34.6 years (range 19-75). We recorded ~2,300 data points per subject to obtain a comprehensive perceptual phenotype, comprising multiple perceptual measures of 66 diverse odours. We show that general olfactory acuity correlates with gender, age, race, smoking habits, and body type. Young, female, non-smoking subjects had the highest average olfactory acuity. Deviations from normal body type in either direction were associated with decreased olfactory acuity. Beyond these factors we also show that, surprisingly, there are many odour-specific influences of race, age, and gender on olfactory perception. We show over 100 instances in which the intensity or pleasantness perception of an odour is significantly different between two demographic groups.

Conclusions: These data provide a comprehensive snapshot of the olfactory sense of a diverse population. Olfactory acuity in the population is most strongly influenced by age, followed by gender. We also show a large number of diverse correlations between demographic factors and the perception of individual odours that may reflect genetic differences as well as different prior experiences with these odours between demographic groups.

Figure 1

Subject population. a, Self-reported demographic, personal habit, and product usage information about the subject population compared (when available) to 2010 US Census data for New York City, as indicated by the red vertical line. b, Self-reported ethnicity of the subject population sorted by self-reported race and listed with the most frequently reported ethnicities at the top. Data for Caucasians are split into two columns, such that “German” is the most frequently and “Native American” the least frequently reported ethnicity. Subjects could self-identify as more than one ethnicity and only ethnicities that were used by at least two subjects are shown. 75% of “Other” subjects self-;identified as Hispanic compared to fewer than 10% in the three main racial groups. c, Primary residence of 93% of study subjects is indicated as a yellow square on a satellite map of the NYC metropolitan area, labelled to indicate the five boroughs, New Jersey, and Long Island. The approximate boundary of Manhattan is indicated by the magenta line. In some cases many subjects were from the same area and their individual locations cannot be resolved here. The remaining subjects did not have their primary residence within the area covered by the map.

Figure 2

Experimental design.a, The study involved four identical visits. 412 subjects completed the first two visits. Of those, the 21 with the lowest general olfactory acuity were excluded (subjects in red). 56 subjects were invited for visits three and four (subjects in green). b, There was a spacing of approximately seven days between visits one and two and between visits three and four. There was a spacing of more than one year between visit two and three. c, During each visit, the intensity and pleasantness of 159 stimuli was rated. After the subjects rated 53 stimuli they assigned descriptors to spearmint oil, pentadecalactone, and the solvent propylene glycol. After they rated 106 stimuli, they assigned descriptors to vanillin and androstenone. After the intensity and pleasantness rating the thresholds for pentadecalactone, isovaleric acid, and vanillin were measured. See the “odours and sequence of stimuli” tab of Additional file 1 for the complete sequence of odours and concentrations used in the study.

Figure 3

Within-individual variability of odour perception. The average absolute change in intensity and pleasantness ratings of fifteen stimuli were measured for 56 subjects twice on each of four visits. a, The absolute intensity and pleasantness changes were calculated by averaging the absolute difference between two ratings performed by the same subject. For the 30 min measure, the first and second ratings were compared for all four visits. For the 7 days measure, the second rating of the first visit and the first rating of the second visit and the second rating of the third and first rating of the fourth visit were compared. For the 1 year measure, the second rating of the second visit and first rating of the third visit were compared. There was about one week between visits one and two and between visits three and four. Visits two and three were more than one year apart. b, The average absolute change in intensity (left) and pleasantness (right) ratings for each stimulus is shown for three time intervals between ratings: 30 minutes, one week, and more than one year (519 days on average). An absolute change of “1” is equivalent to a change of one step on the rating scale (for example from “very weak” to “extremely weak”). Data are plotted as mean±standard deviation of the mean (S.D.). Odour names are labelled according to odour concentration: grey: low; black: high.

Figure 4

General olfactory acuity.a-f, Olfactory acuity ranks for different demographic groups are shown and compared. Each subject was assigned a rank between 1 and 391, with 391 signifying the highest, or best, olfactory acuity. Subjects are represented by grey dots and the median and first and third quartile of each group are shown. A two-tailed Mann–Whitney test (*p<0.05; **p<0.01; ***p<0.001; n.s.: not significant) was performed to test the differences for statistical significance (Ns: < 35 years: 195; > 35 years: 196; Women: 210; Men: 181; Underweight: 28; Just Right: 202; Overweight: 149; Asians: 31; African-Americans: 97; Caucasians: 178; Non-Hispanic: 305; Hispanic: 77; Non-Smokers: 290; Smokers: 92). Subjects were divided by age according to the median of 34.6 years, which is rounded to 35 years for labeling the figure.

Figure 5

Perception of odour intensity.ac, Thresholds for pentadecalactone (a), vanillin (b), and isovaleric acid (c) for different demographic groups. A two-tailed Mann–Whitney test (*p<0.05; **p<0.01; ***p<0.001; n.s.: not significant) was performed to test the differences for statistical significance. Mean±S.D. are shown. dh, Differences in intensity rank of stimuli. Only statistically significant differences are shown. The test for statistical significance was a two-tailed Mann–Whitney test with a sequential Bonferroni correction called the Holm’s method to correct for multiple comparisons (p<0.0082) [47]. Odour names are labelled according to odour concentration: grey: low; black: high; (Ns: < 35 years: 195; > 35 years: 196; Women: 210; Men: 181; Asians: 31; African-Americans: 97; Caucasians: 178; Perfume users: 267; Non-perfume users: 112). Subjects were divided by age according to the median of 34.6 years, which is rounded to 35 years for labeling the figure.

Figure 6

Perception of odour intensity and pleasantness.a-b, 66 odours at two concentrations are ordered according to how intense (a) or how pleasant (b) they are perceived to be by the subject population. Odour names are labelled according to odour concentration: grey: low; black: high. Histograms are shown to allow the evaluation of variability. The inset shows the legend for the histograms. The blue lines show statistically significant correlations between the perceived intensity and pleasantness of a given stimulus.

Figure 7

Perception of odour pleasantness in different demographic groups.a-g, The evaluation presented in a-g of seven demographic groups is derived from the full population in Figure 6. Only the 20 most pleasant and the 20 most unpleasant stimuli are shown for each group. Odour names are labelled according to odour concentration: grey: low; black: high. The inset shows the legend for the histograms (Ns: < 35 years: 195; > 35 years: 196; Women: 210; Men: 181; Asians: 31; African-Americans: 97; Caucasians: 178). Subjects were divided by age according to the median of 34.6 years, which is rounded to 35 years for labeling the figure.

Figure 8

Demographic differences in perceived odour pleasantness.ae, Differences in pleasantness rank of stimuli. Only statistically significant differences are shown. The test for statistical significance was a two-tailed Mann–Whitney test with a sequential Bonferroni correction called the Holm’s method to correct for multiple comparisons (p<0.0082) [59]. Odour names are labelled according to odour concentration: grey: low; black: high; (Ns: < 35 years: 195; > 35 years: 196; Women: 210; Men: 181; Asians: 31; African-Americans: 97; Caucasians: 178; Perfume users: 267; Non-perfume users: 112). Subjects were divided by age according to the median of 34.6 years, which is rounded to 35 years for labeling the figure.

Figure 9

Demographic differences in perceived odour quality.a, Colour code for the categories of descriptors used. b, g, l, The molecular structure of androstenone (b), pentadecalactone (g), and vanillin (l) and the frequency with which descriptors were applied to each odour are shown. Only descriptors that were used in at least 10% of the visits are shown. The frequency of descriptor usage is represented by the area of the rectangles. c-f, h-k, m-p, The percentage differences between demographic groups in descriptor usage for descriptors that were applied to androstenone (1/10,000 dilution) (c-f), pentadecalactone (1/500 dilution) (h-k), or vanillin (1/200 dilution) (m-p) in more than 10% of the visits (Ns: < 35 years: 195; > 35 years: 196; Women: 210; Men: 181; Asians: 31; African-Americans: 97; Caucasians: 178). Subjects were divided by age according to the median of 34.6 years, which is rounded to 35 years for labeling the figure.