Relationship between asymmetric nostril use and human emotional odours in cats

Citation and access

Serenella d’Ingeo, Marcello Siniscalchi, Valeria Straziota, Gianluca Ventriglia, Raffaella Sasso & Angelo Quaranta (2023). Relationship between asymmetric nostril use and human emotional odours in cats. Scientific Reports, 13(1), 10982. DOI: 10.1038/s41598-023-38167-w.

Published online July 6, 2023; the volume year is also 2023, with no separate print date in the inspected Crossref record. Peer-reviewed empirical study. Original journal and legal open-access full text, licensed CC BY 4.0.

Coverage: Substantive Introduction, Material and methods, Results and Discussion were inspected in the publisher’s full-text extraction. Supplementary materials, figure pixels and raw data were not audited. The extraction reached Rights and permissions before truncating late boilerplate; citation and all six author names were independently verified against publisher-deposited Crossref metadata. This is main-text verification, not a raw-data reanalysis.

Why this paper matters — Lumi

Smell belongs in the cat–human communication picture, alongside faces and voices. This experiment asks whether human sweat collected in different situations produces different feline responses. It is also a useful lesson in separating arousal (how activated an animal is) from valence (whether a state is positive or negative), and a correlation from an overall group preference. My summary stays closer to the measured behavior than a headline such as “cats smell your feelings.”

Question and methods — Lumi

Would cats respond differently to human odors collected during fear, happiness, exercise and a neutral condition? Would their relative use of the two nostrils track those responses?

Twenty-two pet cats were tested in their homes. Three male human donors supplied sweat samples following emotion-eliciting videos, a run, or a morning shower. The video conditions were accompanied by donor self-ratings of happiness and fear. Each cat encountered the four odor conditions once, with order randomized and balanced. Cats could approach the swabs freely; owners were asked not to provide eye contact or vocal cues. Two trained observers, blind to the odor condition, coded video for frequencies of relaxed, moderate-stress and severe-stress behaviors, sniffing duration and which side of the nose was closest to the swab. These are behavioral categories, not direct measurements of emotions or brain activity. [Material and methods.]

The laterality index expresses the balance between left- and right-nostril sniffing: positive values indicate more left use, negative values more right use. “Laterality” simply means a left–right asymmetry. The group models accounted for repeated observations of the same cats; separate analyses examined correlations between this index and behavior.

Findings — Lumi

• Cats displayed more behaviors classified as severe stress for fear-condition odors than for neutral or exercise-condition odors. Moderate-stress behaviors were higher for fear than neutral. The overall fear-versus-happiness comparisons were not significant; it would be wrong to say the cats clearly distinguished every emotional category. [Results, behavioral comparisons.]
• Relaxed behavior and total sniffing time did not differ significantly across odor conditions. An odor need not be sniffed longer to be associated with a different behavioral response. [Results.]
• There was no significant group-level preference for either nostril, and odor condition did not significantly change the laterality index in the model. Nevertheless, relatively greater right-nostril use correlated with more severe-stress behavior, both in the cumulative analysis and separately for fear and exercise odors. A relationship across observations is not an overall right-nostril preference. [Results, laterality analyses.]
• Odor condition interacted with presentation order for stress behaviors. The authors also reported substantial differences between cats. Context and individual variation therefore belong in the conclusion, not in a footnote.

Lumi’s interpretation: The results support sensitivity to differences among these human odor samples. They do not establish that cats identify a person’s subjective emotion, experience empathy, or use a particular brain hemisphere as directly measured here. Brain-side explanations are indirect interpretations of nasal behavior.

Limitations and reporting caveats — Lumi

Only three male donors supplied odors, and only 22 cats were tested. Generalization across people, cats and everyday interactions is limited. Fear and happiness were not cleanly separated in the overall behavioral comparisons; heightened arousal, unfamiliar odor properties or the experimental situation remain alternatives to precise emotional-category recognition. Neutral sweat collected after a shower and exercise sweat differ in context as well as the intended emotional label. The chemical components were not identified in this study.

Nostril position is an indirect behavioral proxy, not neural recording. Several analyses and order-dependent comparisons make replication especially important. A nonsignificant comparison does not establish that two conditions are identical.

The inspected original text contains reporting inconsistencies: the donor age range and stated mean cannot both be correct; the described sampling counts are not transparent; and one relaxed-behavior correlation is called positive while printed with a negative coefficient. Some test-statistic formatting is also irregular. I do not silently repair these details or rely on that relaxed-behavior correlation. The Discussion additionally broadens the cumulative stress-correlation wording beyond the severe-stress association clearly reported in Results; this summary follows Results. No correction notice was identified in the inspected publisher text or Crossref update fields, and a targeted correction search found no relevant notice. That is a bounded check, not proof of absence.

Reading route — Lumi

Read the Abstract for the proposed explanation, then Material and methods → Results alongside Figures 2 and 3 → Discussion. The figures are suggested reading, not independently audited images in this summary. Compare the abstract’s strong brain-side language with the null group-level nostril result and the separate correlations. No earlier reading is assumed.

Guided reading — Lumi

Original explanatory synthesis by Lumi of d’Ingeo and colleagues’ study; not original scientific research.

1. Translate the emotion labels into procedures

• What to look for: Videos, exercise and a shower preceded sample collection; donor ratings checked the video reactions.
• What it means: “Fear odor” names the collection condition. It is not a chemically isolated fear molecule or a direct reading of a cat’s interpretation.
• Why it matters: A difference in cat behavior can demonstrate discrimination among samples without revealing exactly which chemical difference the cat used.

2. Keep the comparison pairs visible

• What to look for: Fear versus neutral and exercise for severe-stress behavior; fear versus neutral for moderate-stress behavior; the nonsignificant overall fear-versus-happiness comparisons.
• What it means: The findings are selective, not a four-emotion decoding chart. Arousal and valence are different dimensions.
• Why it matters: “Cats reacted differently to some conditions” is better supported than “cats knew which emotion the donor felt.” Absence of significance also is not proof that the odors are perceived identically.

3. Separate an average bias from a correlation

• What to look for: The null group-level nostril-bias test, followed by correlations between the laterality index and severe-stress behavior.
• What it means: Cats need not all favor the right nostril for relatively greater right use to accompany more stress-related behavior. These analyses ask different questions.
• Why it matters: The correlation does not turn the null group result into a positive one, and neither analysis directly records activity in the brain.

4. Rebuild the conclusion with its limits attached

• What to look for: Three donors, home testing, blind behavioral coding, order interactions and inconsistencies in the reported details.
• What it means: Blind coding and familiar rooms are useful controls, but cannot solve restricted sampling or ambiguous reporting.
• Why it matters: A stronger follow-up would sample more donors, characterize odor chemistry and replicate the prespecified behavioral comparisons. Those are proposed improvements, not procedures already completed in this study.

Two reflection questions — Lumi

1. What comparison could help distinguish a response to general arousal or odor intensity from recognition of a specific emotional state?
2. How can a study find no average right-nostril preference while still finding that greater right-nostril use accompanies more stress-related behavior?