Names Can Change Odor Pleasantness Without Visual Imagery

The same odor can seem pleasant when we hear a positive name and unpleasant when we hear a negative one. Does this change occur because the name makes us picture a food or an object in our mind?

A 2021 study compared people who had been blind since early childhood, people who lost their sight after age 3, and sighted people. Spoken names changed odor pleasantness in every group, with larger changes in the two blind groups. The results make it difficult to regard visual imagery as an essential mechanism behind this phenomenon.

Investigating why names matter by comparing visual experience

The study aimed to do more than demonstrate once again that names can change odor ratings. It investigated what supports that change: the meaning of words, visual imagination, or both.

The researchers predicted that if visual images evoked by words were important, names would have a smaller effect in people with little visual experience from early childhood. This hypothesis and the main analysis plan were registered before the experiment.

There were 64 participants: 17 in an early-blind group who had lost their sight before age 3, 15 in a late-blind group who had lost it after age 3, and 32 age- and sex-matched sighted participants. Thirteen members of the early-blind group were congenitally blind. The study defined blindness as visual acuity below 0.05. Not all 32 blind participants lacked all light perception; 17 were totally blind.

Hearing a recorded name before smelling the same odor

All participants wore blindfolds, heard a recorded name through a loudspeaker, and then smelled an odor in a jar for 3 seconds. Recordings were used to keep the tone of voice consistent across conditions.

There were 10 experimental odors, each presented once with a positive name, once with a neutral number, and once with a negative name. For example, citral was labeled “Squeezed lemons,” “26,” and “Toilet cleaning detergent,” while isoamyl acetate was labeled “Ripe banana,” “35,” and “Paint thinner.” These were different descriptions applied to the same samples.

To make the repetitions less obvious, 15 other odors were interspersed, giving 45 presentations in total. At least five samples intervened before the same experimental odor reappeared. Samples were separated by 60 seconds, and a break was included during the task.

Participants gave verbal ratings of pleasantness, intensity, and familiarity on separate nine-point scales. They were told beforehand to rate the odors themselves rather than the names. They then completed a separate task identifying 16 odors by choosing from four names, and were finally asked whether they had understood the purpose of the experiment.

Pleasantness was highest with positive names, followed by neutral and negative names

Across all participants, mean pleasantness was 5.96 with positive names, 5.25 with neutral numbers, and 4.53 with negative names. The midpoint of the nine-point scale was 5. All three conditions differed significantly, with ratings decreasing from positive to negative names.

This ordering appeared in the early-blind, late-blind, and sighted groups. The differences between positive and negative names, and between positive and neutral names, were also larger in both blind groups than in the sighted group. In contrast, the difference between neutral and negative names did not differ significantly between groups.

This is therefore not quite the same as finding that blind people rated odors more favorably overall. Pleasantness averaged across all conditions did not differ significantly between groups; what differed was how far the ratings shifted with the names. This was the opposite of the researchers’ prediction that the early-blind group would show smaller changes.

What about intensity, familiarity, and odor identification?

Names also affected intensity, but less strongly than pleasantness. Overall, odors with positive names were rated slightly more intense than those with neutral numbers. The preregistered prediction that negative names would produce the highest intensity ratings was not supported.

Averaged across names, the early-blind group rated odors as more intense than the sighted group. However, the study did not measure detection thresholds, so higher ratings do not establish that participants could detect more dilute odors.

Familiarity was higher with positive than with negative names. There was no clear evidence that the effect of names on familiarity differed by visual-status group. In all three labeling conditions, higher familiarity ratings were associated with higher pleasantness ratings.

On the 16-odor identification task, the early-blind group averaged 12.7 correct answers, the late-blind group 13.1, and the sighted group 12.4, with no significant group difference. At least for this task, the results do not explain the larger name-induced changes in pleasantness simply in terms of better or worse identification ability.

Names still had an effect among those who noticed the experiment’s purpose

Of the 64 participants, 34 had partially or fully noticed repeated odors or changes in their names. When these 34 were analyzed alone, the effect of names on pleasantness and the group differences in that effect remained. The effect on familiarity also remained, but the small effect on intensity was no longer significant.

The authors considered it difficult to explain the naming effect simply as a result of participants failing to notice the experimental manipulation. Still, the questions asked afterward could not eliminate every influence of expectations or response strategies. This is best read as an additional analysis checking the robustness of the results.

Visual imagination alone cannot explain the findings

The authors’ central conclusion was that visual processing itself cannot provide a general explanation for verbally induced changes in olfaction. Names substantially changed pleasantness even in the early-blind group, whose visual experience was limited.

Possible reasons for the larger changes in the blind groups include a greater everyday reliance on verbal descriptions, mental images involving sound, touch, or smell, and memories. However, the study did not directly measure the vividness of mental imagery or how much participants sniffed during the experiment.

Nor did the researchers experimentally manipulate when people lost their sight: they compared people with different experiences. These results alone cannot establish that blindness itself caused larger changes, or that visual imagination plays no part at all in sighted people. The clear clue is that word meanings can enter into odor judgments without visual images being essential.

Summary

In the early-blind, late-blind, and sighted groups alike, the same odors were rated more pleasant with positive names and less pleasant with negative names. In comparisons involving positive names, both blind groups showed larger changes than the sighted group. The prediction that limited visual experience would weaken the effect was not supported.

Odor identification showed no clear group differences, and the effects on intensity and familiarity took different forms from those on pleasantness. The study suggests that giving meaning to a scent depends not only on picturing an object, but also on connections with words and experience.

The research was supported by a Swedish foundation for the humanities and social sciences and by the Swedish Research Council. The funders had no involvement in the research beyond financial support, and the authors declared no conflicts of interest.

References

Cornell Kärnekull S, Gerdfeldter B, Larsson M, Arshamian A. Verbally Induced Olfactory Illusions Are Not Caused by Visual Processing: Evidence From Early and Late Blindness. i-Perception. 2021;12(3). doi:10.1177/20416695211016483