Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611
Performing Department
(N/A)
Non Technical Summary
Consumers' strong desire to indulge in unhealthy foods, and consequent overconsumption of these foods, carry significant negative impacts on public health, healthy food marketing, and sustainability of the food system. Soft-end policy measures, namely nudges, have shown more promising potential than hard-end policies in promoting healthy food decisions. However, the efficacy of nudges in eliciting and maintaining healthy food choices is limited by consumers' strong motivations to consume unhealthy foods, which is driven mainly by the strong positive consumption experience associated with such foods. It is therefore necessary to design behavioral nudges that can shift consumers' perceptions of the tastiness of healthy foods, thus addressing one of the root causes of unhealthy eating. The proposed project utilizes two well-established biases in the experimental economics literature, attribution bias and projection bias, to investigate how induced positive consumption experience and memory with a healthy food impact consumers' preferences for this product when making current and future decisions, as well as choices between the healthy product and an equivalent, unhealthy alternative. We adopt a neurobehavioral approach to improve understanding of the mechanism (attention vs. motivation) through which induced positive taste perceptions increase consumer preferences for healthy foods. By determining the neurobehavioral factors elicited by an induced positive consumption experience with a healthy food, this project will provide a strong conceptual framework for the subsequent design of effective behavioral nudges that can deliver sustained improvements in consumers' dietary choices. This project therefore provides benefits to policymakers, food marketers, and the general public.
Animal Health Component
20%
Research Effort Categories
Basic
60%
Applied
20%
Developmental
20%
Goals / Objectives
This project utilizes two well-established biases in the experimental economics literature, attribution bias and projection bias, to investigate how induced positive consumption experience and memory with a healthy food impact consumers' preferences for this product when making current and future decisions, as well as choices between the healthy product and an equivalent, unhealthy alternative. We adopt a neurobehavioral approach to improve understanding of the mechanism (attention vs. motivation) through which induced positive taste perceptions increase consumer preferences for healthy foods. The following objectives will be pursued:Aim 1: Determine the effectiveness of utilizing attribution bias to promote higher current preferences for healthy food products.Aim 2: Determine the effectiveness of utilizing projection bias to promote higher future preferences for healthy food products.Aim 3: Determine the applicability of attribution and projection bias in nudging mechanisms to shift consumer preferences in favor of healthy food options compared to equivalent, unhealthy alternatives.
Project Methods
The proposed project activities will be carried out over four experimental studies that will be used to collect behavioral and psychophysiological outcomes to determine the effect of induced positive consumption experience and memory with a healthy food product on subsequent preferences for this product when making current and future decisions, as well as when choosing between the healthy product and an equivalent unhealthy alternative.Overview of Experimental DesignIn the experimental studies, participants will be randomly assigned to a control and treatment groups, where each will participate over 2 days. On day 1, subjects in both groups will consume a healthy snack product, however, those in the treatment group will be induced with a positive consumption experience with the healthy snack, while those in the control group will not. The positve consumption experience inducement will be hunger-based, where subjects in the treatment group will consume the healthy snack under a hunger state, while those in the control group will consume the snack under a neutral (not hungry) state. Hunger manipulation will be achieved by asking all subjects to fast for 5 hours prior to their participation in day 1 (to balance initial hunger levels upon arrival) and providing subjects in the control group with a pre-snack (to offset their hunger levels and put them in a neutral state) before consumption of the healthy snack. Subjects will then participate in a cue exposure task and a preference elicitation task to measure their behavioral and neurophysiological indicators of preferences. In the cue exposure task, subjects will be presented with non-target (non-food) and target (healthy snack they consumed) images in a random order. Psychophysiological outcomes related to attention, motivation, and approach/withdrawal will be collected using an electroencephalogram (EEG) to measure electrical brain activity and an eye-tracker to measure pupil dilation. In the preference elicitation task, subjects will report their willingness-to-pay (WTP) for the healthy snack in an incentive-compatible setting, where there is a probability of making a purchase based on their decision.In the first experimental study, the cue exposure and preference elicitation tasks will be completed on day 2 to investigate attribution bias, which is the tendency to rely on past consumption memory when making current decisions. In the second experimental study, the cue exposure and preference elicitation tasks will be completed on day 1 to investigate projection bias, which is the tendency to rely on current consumption experience when making decisions about the future. The third and fourth experimental studies will replicate experiments 1 and 2, respectively, with the additional layer of including an equivalent unhealthy alternative to the healthy product that subjects consumed in day 1. This will assess the effectiveness of utilizing attribution and projection bias to shift consumer taste perceptions and preferences away from unhealthy products and towards healthier alternatives. The cue exposure in experiments 3 and 4 will additionally include images of the equivalent unhealthy alternative (in addition to the target and non-target images). The preference elicitation task will also be replaced with a choice task that asks subjects to choose the product they prefer between the healthy product they consumed in day 1 and the equivalent, unhealthy alternative.Data AnalysisTo confirm success of our positive consumption experience inducement procedure, we will first compare subjects' reported hunger levels between control and treatment immediately prior to consumption of the healthy product. Subjects in the treatment are expected to report higher hunger levels compared to the control. The subjects' reported enjoyment levels during consumption of the healthy product will also be compared between control and treatment. We expect subjects in the treatment group to report higher enjoyment levels compared to the control.The treatment effect of inducing a positive consumption experience/memory with the healthy product on subjects' preferences for the product will be evaluated using behavioral and psychophysiological outcomes. Subjects' WTP and choices for the healthy product will be used as the behavioral outcome and compared across control and treatment. We expect average WTP and choices for the healthy product to be higher in the treatment than the control.Multiple psychophysiological outcomes will be collected from the eye-tracker and EEG headset during the cue exposure stage. The EEG data will be used to measure Event-Related-Potentials (ERPs). The ERP waves will be examined at two time windows post stimulus onset to evaluate two ERP components. Specifically, the P200 (200-300 milliseconds) and P300 (300-400 milliseconds) components will be measured to examine motivation and attention, respectively. To extract the ERP components, the EEG data will be segmented into 2.5-second epochs (one for each image) spanning 800 milliseconds pre-stimulus and 1700 milliseconds post-stimulus. Correction for eye-movement and blinking will be performed following the Gratton and Coles algorithm (Gratton et al., 1983), after which EEG activity above + or - 100 microvolts will be removed to further correct for artifacts (e.g., head or body movements). The pre-stimulus data will be used for baseline correction and the resulting EEG data will be averaged separately over the target (food) and non-target (non-food) stimuli for each subject to generate the ERP components. The difference in those ERP components between the target and non-target stimuli will then be calculated and compared between control and treatment to examine subjects' attention and motivation to the images of the healthy food product. We expect subjects in the treatment group to exhibit higher P200 and P300 waves in response to images of the healthy food product, indicating higher motivation and attention towards the product.The EEG data will also be used to measure brain oscillations, which will be used to calculate the Frontal Asymmetry Index (FAI). The FAI measures differences in alpha wave activity between the left and right sides of the prefrontal cortex (PFC) and indicates approach (higher FAI/higher left PFC activity) or withdrawal (lower FAI/higher right PFC activity) behavior in response to a product stimulus. To extract the FAI, the EEG data will be filtered using a 0.1-40 Hz band-pass filter, after which each epoch will be Fast Fourier transformed to generate the alpha power band (7.75-13.75 Hz). The FAI will be calculated by subtracting the natural log of alpha power between the right and left PFC regions. The difference in FAI between the target and non-target stimuli will be calculated and compared between the control and treatment conditions. We expect subjects in the treatment group to exhibit higher FAI towards the images of the healthy food product, indicating higher approach behavior towards this product.Pupil dilation will also be measured in millimeters using the eye-tracker. More dilated pupils indicate higher approach/arousal towards a product stimulus. The difference in pupil dilation between the target and non-target stimuli will be compared between the control and treatment conditions. We expect subjects in the treatment group to exhibit higher pupil dilation in response to the target stimuli, indicating higher approach/arousal towards the images of the healthy food product.To measure the treatment effects, parametric (t-test) and nonparametric (Mann-Whitney test) comparison tests will be used to examine differences in outcome variables across the control and treatment. Additionally, regression analysis will be conducted to ensure robustness of the findings, and to control for important sociodemographic and behavioral factors.