Skip to content Skip to navigation

Experiments with Appliance Choice

This work showed that even simple neuroimaging experiments can predict large-scale behavior, suggesting that neuroimaging can pilot new eco-labeling and energy information formats to accurately forecast their effect on the general population before implementation.
Year Started: 

This study examined whether some groups of consumers alter their decision-making process when buying major household appliances due to behavioral nudges, such as eco-labeling. Researchers then studied neural activity in each group via brain imaging with functional magnetic resonance imaging. 

In a nationwide survey of appliance purchases by 1,550 people, this research identified differences in consumer responses to the Energy Star label. They characterized the differences the cause overinvestment (pro-environmental attitudes) and underinvestment (high numeracy) in energy efficiency, compared to economically-rational benchmarks.

Using functional magnetic resonance imaging, researchers examined deciding on the purchase of CFL bulbs with different product attributes. They discovered that the Energy Star label reliably activates the nucleus accumbens and caudate, parts of the brain’s “reward center,” in a fashion similar to low prices. The activity in these reward regions predicted CFL purchases not only within the neuroimaging group, but was also able to accurately and significantly forecast the CFL purchases of the 1,550-person nationwide sample. Activity in the medial prefrontal cortex, which is associated with subjective value calculation and cost-benefit assessment, predicted refraining from purchase. This may indicate that decisions which elicit more reflection upon trade-offs may also be associated with more conservative purchasing behavior.

This project also found that:

  • Stronger pro-environmental attitudes were associated with a stronger response in reward regions to the Energy Star label.
  • Individuals’ temporal discounting rates were associated with different neural regions predicting the purchase choice. The decisions of individuals with high discount rates (i.e., more impulsive) were predicted by emotional regions (positively by nucleus accumbens and caudate, and negatively by the anterior insula, associated with negative emotion processing and emotional salience more broadly). Furthermore, these regions and the MPFC were all more active in high discounters when they saw the Energy Star label, indicating that high discounters found the Energy Star label extremely salient compared to those with lower discount rates. In contrast, decisions of individuals with low discount rates were negatively predicted by the MPFC.
  • Combining the national predictions based on neural regions with the individual differences on discount rate, we can infer that the general population is making decisions as if they are using the same regions as individuals with high discount rates. This aligns with energy economists’ analyses of market data which indicate that real-world purchases are made as if much of the population has a very high implicit discount rate.

The researchers are seeking opportunities to study whether information overload plays a role in the Energy Star label’s salience and whether we can optimize Energy Guide labels to improve energy efficiency investment.

Publications: The Heterogeneous Effects of Eco-labels on Internalities and Externalities In prep. Online at Stanford Law School Sahoo, A., Sawe, N. (2015)

Negative Dividends: Internality Losses Can Outweigh Externality Gains International Association for Energy Economics: European Energy Conference Best student paper award. Sahoo, A., Sawe, N. (2014)

Video: Your Brain, The Environment and Our Decisions TEDx Stanford Sawe, N. (2015)