Proof of concept study: Testing human volatile organic compounds as tools for age classification of films.
Level 4 - case-series / case-control
Observational proof-of-concept study evaluating audience emissions without clinical or randomized controls.
PubMed 30307954 · doi:10.1371/journal.pone.0203044
What was done
Airborne volatile organic compounds (VOCs) and carbon dioxide emitted by cinema audiences watching films across different age classifications (0, 6, 12, and 16) were measured. Time-independent features were extracted from the time series data of each compound and evaluated using a random forest model to test their ability to predict age ratings.
What was found
The abstract reports no numerical metrics or classification accuracy values. Most measured VOCs and CO2 were unable to reliably predict film age classifications across categories. However, isoprene emissions reliably predicted age classifications of 0, 6, and 12 across various genres and audience age groups, though they were not predictive for the 16 rating.
Why it matters
This proof-of-concept study suggests that physiological emissions such as isoprene could potentially provide an objective crowd-based metric to assist or complement film rating boards.
Limits
The abstract omits sample sizes for the number of audience members, screenings, and films evaluated, as well as quantitative model metrics. Most compounds failed to predict ratings, the approach was ineffective for the highest rating category (16), and physiological arousal does not capture editorial rating criteria such as language or drug references.