Neurodynamic responses to repeated visual stimuli: an EEG study of food habituation

This research investigates food habituation towards high, low-calorie and non-food images through the repeated presentation of visual stimuli using electroencephalogram (EEG) data analysis. We used well-developed EEG signal analysis techniques to show that the obtained EEG habituation marker reflects different calorie and non-food variations. Furthermore, the habituation rate is slower for high-calorie images than for low-calorie and non-food images. We begin by explaining the overall framework, which enables us to identify good EEG markers for food habituation, perform different analyses, and compute habituation rates for different energy-value foods. The work starts with data acquisition, and following this, we performed data pre-processing, feature extraction and finally, statistical evaluation of the obtained results.

We first focused on comparing behavioural responses to the repeated presentation of the same and varied food images. To achieve this objective, the effects of repetition of the same food image (high-calorie, low-calorie or non-food image) were compared with previous studies that used intermixed images in their experiments. Moreover, the effect was compared on the modulation of early and late components of event-related potential (ERP) in different regions of interest (ROIs) of the brain. Statistical validation revealed that this study's early and late ERP component results could be compared with previous studies where a variety of images were used. It suggests that the effect of repetition of food and non-food images can be obtained by the presentation of a single image repeatedly rather than using a variety of images.

Next, we used the same image repetition paradigm to assess the cognitive processes in different ERP windows and determine how repetition affects habituation in high/low-calorie food and non-food visual stimuli. We split each image category session into five trial groups (in one session, totalling 30 trials with six trials in each group of 24 participants) and averaged over time for each participant. The cluster-based statistical test revealed that the significant main effects and interactions were observed in low-calorie and non-food image in the parietal area of the brain (around 165 ms - 300 ms time window) but not in the high-calorie image. The reason is likely due to high-calorie images showing sustained attention compared to low-calorie and non-food images. Additionally, the habituation rate was faster for non-food images, followed by the low and high-calorie images.

Furthermore, we used the same image repetition paradigm to assess the cognitive processes in different time windows and in different frequency bands and determine how repetition affects habituation in high/low-calorie food and non-food visual stimuli. We split each image category session into six trial groups (in one session, totalling 30 trials, so five trials in each group) and averaged over time in different frequency bands for each participant. The cluster-based statistical test revealed that the significant main effects and interactions were observed in non-food images in the frontal theta band (4 - 7 Hz) with a time window range of 110 ms - 330 ms. However, no statistical significance was identified for either high or low-calorie food images. This could be due to food images showing sustained attention compared to non-food Images.

Another study examines how habituation and dishabituation influence brain responses to repeated presentations of food and non-food images across different time windows. Using EEG-derived ERPs, the study analysed neural adaptation to high-calorie, low-calorie, and non-food stimuli, focusing on changes in attentional and motivational engagement over repeated exposures. The results revealed significant habituation effects for non-food images within the 160 ms - 290 ms time window over the parietal brain region, while food images, particularly high-calorie ones, maintained prolonged neural engagement. Additionally, introducing novel stimuli led to response recovery, demonstrating dishabituation effects across all image categories. These findings highlight the role of food-related attentional biases in eating behaviours and suggest potential interventions, such as modifying food exposure patterns, to regulate energy intake and promote healthier dietary habits.

Furthermore, in the functional connectivity analysis study, theta-band connectivity patterns revealed significant habituation effects for non-food images, while high-calorie and low-calorie foods sustained prolonged neural engagement. The connectivity matrix analysis demonstrated a decline in frontal-parietal connectivity for non-food images, whereas high-calorie and low-calorie foods maintained stable connectivity across trials. The habituation rate analysis confirmed a rapid decline for non-food stimuli, a moderate decline for low-calorie foods, and minimal habituation for high-calorie foods, highlighting their persistent attentional salience. These findings suggest that functional connectivity measures can serve as objective markers of food-related attentional biases, offering insights into cognitive mechanisms that influence eating behaviours and potential strategies for dietary interventions.

Finally, we examined the neurophysiological mechanisms of attentional habituation within a session and its relationship with Body Mass Index (BMI) during repeated exposure to high-calorie, low-calorie, and non-food images. ERPs were analysed in both high and low BMI groups to assess changes in attentional engagement over repeated exposures. The findings indicate significant habituation effects in the parietal brain region for non-food images, with noticeable declines in attention and motivation within the 170 ms - 310 ms and 180 ms - 320 ms time windows for low and high BMI groups, respectively. However, no significant habituation effects were observed for high-calorie or low-calorie food images in either BMI group. Furthermore, high-calorie images exhibited the slowest habituation rate, followed by low-calorie and non-food images. This suggests that an individual's brain stays focused on food images for a longer time, whether they are overweight or not. For those with a high BMI, this could make them more likely to continue gaining weight, and for those with a low BMI, it may increase the risk of becoming overweight or obese in the future.

Overall, this research provides strong evidence that high-calorie food images sustain prolonged neural engagement compared to low-calorie and non-food images, as indicated by EEG-derived ERP analyses, functional connectivity measures, and habituation rate assessments. The findings suggest that food-related attentional biases are stronger for high-calorie stimuli, which may play a crucial role in influencing eating behaviours and dietary choices. The slower habituation rate for high-calorie foods implies that repeated exposure does not lead to rapid cognitive disengagement, which may contribute to difficulties in regulating energy intake and a heightened risk of overeating.

A key takeaway from this research is that early neural processing, as reflected in ERP components, along with activity in the frontal and parietal brain regions and theta-band oscillations, serve as critical EEG markers of habituation. Significant habituation effects were observed for non-food and low-calorie images in these regions and frequency bands, whereas high-calorie images maintained stable connectivity and prolonged engagement. This suggests that attentional disengagement occurs more readily for non-food and low-calorie stimuli, while high-calorie foods trigger sustained attention, likely due to individuals' attention and motivation towards high-calorie food. These findings have broader implications for understanding the cognitive and neural mechanisms that influence dietary choices and food-related self-regulation.