Permanent link for Research Review: "How to accurately measure attention to video advertising" on January 8, 2025

A 2024 study is making waves in attention studies, studying what biometric measures most accurately measure video advertising attention. 

Attention-based analytics are becoming increasingly popular amongst marketers and advertisers, but with a lack of industry standardization for testing procedures, attention metrics have very little comparable validity. This study determined the most accurate biometric measures for attention using iMotions software, comparing each measure to EEG results as a “gold standard”.

How was this study conducted:

1. A pre-test was conducted where participants self-reported how much of their attention was captured from 0-100%. Out of the 100 ads shown, the authors selected 10 ads with very high and low attention ratings.
2. Participants were shown a 4 minute relaxing zen garden video to collect their baseline measurements.
3. Participants watched 60 minutes of television interrupted with 5 ad breaks of 6 ads apiece. Of the 30 ads shown, 10 ads were tested and 20 ads were filler.
4. Assorted physiological measures were tracked such as heart rate, eye tracking, blink duration, fixations per second, individual attention dispersion, skin conductance, and facial expression. These measures were tracked against a “gold standard”: EEG, which “is typically not commercially scalable because it requires measurement in labs but was sufficiently scalable and affordable for the purposes of this study. Lower frequency EEG alpha brain waves indicate low attention; therefore, an EEG signature of attention is alpha dipping below the baseline level (Kolar et  al. 2021; Reeves et  al. 1985)” (Harnett et al., 2024).

Findings:

The researchers found that heart rate through BPM is the most accurate measure of attention. Measuring heart rate with BPM is also linked to conscious processing, which the researchers concluded was indicative higher attention. In the past, researchers criticized self reporting of attention since it doesn’t capture unconscious processing, but the authors argue in this paper that self-reporting is valuable because it captures conscious attention. For this reason, biometric measures such as eye-tracking may not be as useful as we previously thought for tracking attention in video advertisements. This is due to the fact that automatic visual processing can do a lot of heavy lifting while a subject is failing to consciously process anything that is happening on screen. Additionally, measures like eye-tracking don’t account for accompanying audio stimuli. These ideas are backed by the data from this study, where perceived “attention” through eye-tracking don’t correspond to the heart rate BPM measurements or the “gold standard” EEG measurement. 

What does this mean for us?

1. In the next few years, heart rate BPM measurements may emerge as an industry standard practice for measuring attention in video advertisements.
2. Tracking BPM is quite an accessible measurement, meaning we may see an increase in companies using biometric studies to inform their market research and advertisement creation.
3. As advertisers, we should be on the lookout for more research on conscious vs unconscious processing in video advertisements, to further indicate the effects of each on consumers.

Fun facts about the study:

1. The researchers tested heart rate, eye tracking, blink duration, fixations per second, individual attention dispersion, skin conductance, and facial expression.
2. Ideally, fMRI would be used as a golden standard to compare each biometric variable, but since it isn’t scalable, EEG was used instead.
3. The study started with 324 compiled 30 second advertisements, which was narrowed down to 100 for testing, before landing on the 10 used in the study.
4. The ads were from the following categories: “financial services, Internet service providers, online retail, skin care, automotive, whitegoods, electronics, snacks, side dishes, and household cleaners”, and the authors made sure each ad was from a different brand in order to test brand recall.
5. There were 261 participants in the study, resulting in 181 complete data sets.

 

Click here to read the full article.

Hartnett, N., Bellman, S., Beal, V., Kennedy, R., Charron, C., & Varan, D. (2024). How to accurately measure attention to video advertising. International Journal of Advertising, 1–24. https://doi.org/10.1080/02650487.2024.2435164

Categories: Biometrics Research
Posted by Georgia Hessel on Permanent link for Research Review: "How to accurately measure attention to video advertising" on January 8, 2025.