Can Fragranced Cosmetics Change Impressions of Personality?

Can wearing fragranced cosmetics change not only how pleasant someone smells but also the impression of their personality? In a 2016 study, body odor collected while people used their usual cosmetics was rated as more pleasant and attractive than body odor collected without them. At the same time, some personality ratings inferred from odor became somewhat more distant from the donors’ own self-ratings.

The study compared how people viewed themselves with the kind of person others imagined them to be based on smell alone. It examined fragrance-based impression formation in greater detail than simply asking whether someone seems like a nice person.

Collecting samples from 113 people with and without cosmetics

There were 113 odor donors: 58 women and 55 men. They wore cotton pads under their arms for 12 hours to collect samples under two conditions. First, they used unscented soap and avoided fragranced cosmetics. Next, they were allowed to use their usual fragranced cosmetics. They were instructed to avoid strongly scented foods, alcohol, and similar influences.

The types and amounts of products were unrestricted. The design aimed to approximate everyday use, including perfumes, deodorants, and antiperspirants. As a result, it did not separate the effects of particular fragrances from those of non-fragrance ingredients.

The donors completed a brief personality questionnaire and items concerning dominance. Neuroticism here is a personality measure concerning susceptibility to anxiety and emotional upset, not a diagnosis of illness. Dominance is likewise a questionnaire measure of the tendency to take the lead in relationships with others.

The raters were female students. The original paper gives different numbers: 67 in the abstract and 68 in the Methods. Each rater smelled 12 samples from six donors, one sample with and one without cosmetics from each donor, and rated the personality they imagined from the odor using the same scales as the donors. They subsequently rated odor intensity, attractiveness, and pleasantness.

Pleasantness and attractiveness increased, but average personality ratings did not change uniformly

Odor pleasantness and attractiveness were significantly higher with cosmetics than without them. The difference in intensity between conditions was not clear.

For average personality ratings, cosmetics use had no significant overall effect on agreeableness, dominance, neuroticism, or extraversion. Extraversion did, however, show a difference by donor sex: for female samples alone, donors were rated as more extraverted when they used cosmetics.

In other words, making an odor more pleasant did not move every personality impression in a desirable direction. The findings depended on the characteristic being assessed.

The gap from self-ratings widened for some traits

The researchers examined agreement between odor-based judgments and donors’ self-ratings in two ways. The first was the distance between scores. For example, if a donor rated themselves 5 and a rater gave them 7, the gap was two points.

This gap was significantly larger with cosmetics for dominance and neuroticism. There was no clear difference between conditions for agreeableness or extraversion. This indicates that even when average personality ratings do not change, their correspondence with individual self-ratings can change.

However, the reference point for agreement was the donor’s own report. The study did not test whether an objectively true personality could be diagnosed from smell; it measured the closeness between self-image and others’ impressions.

Rank correlations call for a more cautious interpretation

The second method used rank correlations to ask whether people with higher self-ratings also received higher odor-based ratings. Without cosmetics, the correlations were 0.20 for neuroticism and 0.34 for dominance, both small but significant positive associations. With cosmetics, these became 0.15 and 0.21, respectively, and only dominance retained a significant association.

Importantly, when the correlation coefficients themselves were compared between conditions, their differences were not significant. An association being significant in one condition but not in the other does not, by itself, establish that the correlation reliably weakened.

The authors emphasized the possibility that cosmetics alter cues to neuroticism while cues to dominance remain. An accurate summary therefore needs two parts: score differences from self-ratings increased for dominance and neuroticism, but a clear between-condition decrease in rank correlations was not established.

Several mechanisms could change impressions

The authors discussed the possibility that people choose fragrances to present themselves favorably, or that odor pleasantness evokes personality associations. However, participants were not instructed to deliberately change the personality impression they conveyed.

The rating method was also examined in advance. A separate group of 56 people was divided into two groups to compare rating personality items consecutively for each odor with rating all samples separately for each item. Because there was no clear difference in correlations among the items, the main study used consecutive ratings. This checked the procedure, but did not prove that the halo effect, in which one impression spills over into other judgments, was entirely absent.

The main study also used a fixed order, from no cosmetics to cosmetics use, and all raters were women. The precision of the short personality questionnaire, gender stereotypes evoked by odors, and differences in product types and amounts could also influence the results. The study does not yet provide a practical formula for using a favorite perfume to project a particular personality.

Summary

Body odor collected during usual use of fragranced cosmetics was rated as more pleasant and attractive than odor collected without them. Average personality ratings did not clearly change overall; only female donors’ extraversion ratings were higher with cosmetics.

Meanwhile, the gap between odor-based judgments and donors’ self-ratings widened for dominance and neuroticism. Between-condition differences in rank correlations were not clear. Fragrance may affect not only pleasantness but also the correspondence between the impressions imagined about a person and their self-image. This study illustrates that complexity.

The paper reports support from Poland’s Ministry of Science and Higher Education and National Science Centre, as well as Czech Science Foundation support for an author. The authors declared no conflict of interest.

Reference

Sorokowska A, Sorokowski P, Havlíček J. Body Odor Based Personality Judgments: The Effect of Fragranced Cosmetics. Frontiers in Psychology. 2016;7:530. doi:10.3389/fpsyg.2016.00530