www.ijcrsee.com
15
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Original scientific paper
Received: February 27, 2025.
Revised: April 18, 2025.
Accepted: April 22, 2025.
UDC:
159.91
612.82-008.6
10.23947/2334-8496-2025-13-1-15-31
© 2025 by the authors. This article is an open access article distributed under the terms and conditions of the
Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
*
Corresponding author: edenisova@donstu.ru
Abstract: The article explores the characteristics of evoked brain activity during food preference decisions, emphasiz-
ing the role of psychological and neurophysiological mechanisms. The relevance of studying eating behavior as a multifaceted
phenomenon is highlighted, with attention to the cognitive, emotional, and physiological factors that influence food preferences.
The study involved 40 participants (70% female). Psychological testing included the Dutch Eating Behavior Questionnaire (Rus-
sian version by I.G. Malkina-Pykh, 2007), the Thought and Behavior Questionnaire (adapted by A.V. Anikina and T.A. Rebeko,
2009), the Three-Factor Eating Questionnaire (Russian version, 2018), and the General Nutrition Knowledge Questionnaire
(translated version of Kliemann, 2016). Neurophysiological data were collected using EEG tasks based on a Go/NoGo paradigm.
Mathematical and statistical methods included the Shapiro-Wilk test, Mann-Whitney U test, Student’s t-test, and k-means cluster
analysis. The study revealed significant differences in brain activity between groups with varying psychological characteristics
and levels of nutritional knowledge. These findings align with previous research, confirming the link between cognitive control,
impulsivity, and food preferences. Enhanced activation in the temporo-occipital regions was observed in participants with higher
nutritional awareness. The role of psychological traits was found to outweigh knowledge levels in shaping dysfunctional eating
patterns, highlighting the need for individualized approaches in prevention and treatment. The limitations, including the sample
size and absence of participants with clinical eating disorders, are discussed alongside recommendations for future research.
Keywords: eating behavior, rational consumption, cognitive control, evoked activity, ERP, Food Preference.
Pavel Ermakov
1, 2
, Ekaterina Denisova
1*
, Daria Kirpu
1
, Anastasia Gosteva
1
, Nadejda Sylka
1
1
Department of Psychophysiology and Abnormal Psychology, Don state technical university, Rostov-on-Don,
Russian Federation, e-mail: paver@sfedu.ru, edenisova@donstu.ru, d.kirpu@list.ru, agosteva@donstu.ru, gramtysh99@gmail.com
2
Regional Scientific Center of the Russian Academy of Education, Southern Federal University, Rostov-on-Don,
Russian Federation,
e-mail: paver@sfedu.ru
Evoked Brain Activity in Food Preference Decisions: Links to Eating
Behavior and General Nutritional Knowledge
Introduction
In recent years, research on the psychophysiological mechanisms of eating behavior has expand-
ed into several distinct directions. Numerous studies highlight distinctive patterns of brain activity in indi-
viduals with eating disorders, with meta-analyses identifying key research areas (Wonderlich et al., 2021).
One of these focus areas includes examining cognitive control and reward system mechanisms through
neuroscience-derived approaches.
fMRI studies have shown the significance of the reward system in regulating eating behavior. Altera-
tions in the reward system have been observed in individuals with compulsive overeating, characterized by
heightened responsiveness to food stimuli. This sensitivity is associated with stereotypical behavioral patterns
and increased susceptibility to external food cues (Leenaerts et al., 2022; Vrieze and Leenaerts, 2023)
. It was
also shown that the activation patterns of the reward system in response to food stimuli can predict not only
food preferences but also Body Mass Index (BMI) (Cosme and Lopez, 2020). These findings are supported
by twin studies, which reveal that childhood impulsivity and reward system activity correlate with BMI in
adulthood. This highlights the long-term influence of early reward system functionality on eating behavior
(Kan et al., 2020). Variations in reward processing and cognitive control play a significant role in shaping
food preferences and obesity risk. Dysfunctions in the orbitofrontal cortex, which integrates food reward
signals, have been linked to heightened responsiveness to food cues and overeating (Rolls, 2021).
www.ijcrsee.com
16
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Building on the understanding of the reward system’s role in eating behavior, cognitive control has
also emerged as a critical factor influencing maladaptive weight-loss strategies in eating disorders. Deficits
in cognitive control are closely linked to impulsivity, a key predictor of pathological overeating (Пичиков et al.,
2018; Oliva et al., 2019). Studies demonstrate that individuals with compulsive overeating exhibit reduced
inhibitory control and attentional selectivity, which also impacts cognitive task performance. For instance,
differences in brain activity during Go/NoGo tasks have been observed predominantly in individuals with
severe disorders manifestations (Berner et al., 2023; Lyu et al., 2018). Moreover, cognitive control levels
have been shown to predict treatment outcomes in cognitive-behavioral therapy (CBT) for eating disorders,
particularly binge eating and obesity, as well as the durability of remission. For example, individuals strug-
gling with interference control in the Stroop task exhibited lower treatment efficacy (Hamatani et al., 2021).
ERP studies have highlighted distinct neural correlations of eating behavior in individuals with
excess weight, particularly the P300, N200, and P200 components. These components reflect variations
in cognitive control, behavioral regulation, and appetite response. A systematic review (Chami, 2019) of
ERP studies of neural responses to food and non-food stimuli among individuals with eating and weight
disorders revealed an enhanced attentional bias toward food stimuli, as evidenced by elevated P300 and
LPP amplitudes, across groups of individuals with excess weight from those of healthy weight. However,
among individuals with obesity, the N200 amplitude positively correlated with caloric intake, while the
P300 amplitude was sensitive to hunger levels (Chami, 2019).
P300 amplitude increase in response to food stimuli, regardless of type, has been observed in groups
with excess weight. Interestingly, the presence of this component, coupled with a preference for low-calorie
food in experimental settings, also predicted snack frequency within the sample. Similarly, the emergence of
the N200 component has been associated with low levels of eating behavior control in obesity (Biehl et al.,
2020). The P200 component has also been highlighted in several studies, showing increased prevalence in
individuals with excess weight in response to food stimuli (Liu et al., 2020; Schwab et al., 2021). This com-
ponent has been linked to appetite suppression when a discrepancy arises between expected and actual
food stimuli (Schwab et al., 2021). However, the relationship between cognitive control and obesity remains
ambiguous, as some animal studies suggest that rather than cognitive control deficits causing obesity, obe-
sity itself might lead to diminished cognitive control (Davidson et al., 2019). At the same time, lifestyle factors
may amplify or modify these neural predispositions. Research suggests that diets rich in ultra-processed
foods not only distort reward system functioning but also reinforce habitual overeating, creating a feedback
loop that exacerbates obesity risk (Edwin and Tittgemeyer, 2020). Moreover, the interaction between be-
havioral patterns and reward system activity may perpetuate maladaptive eating behaviors. For instance,
repetitive exposure to specific food-related cues can alter the reward system’s sensitivity, maintaining dis-
ordered eating tendencies and shaping long-term dietary habits (Frank et al., 2021). So, these findings may
also suggest that while neural activity in response to food cues is heightened in individuals with eating and
weight disorders, its manifestation may depend on specific states (e.g., hunger) or traits (e.g., restrained
eating). Furthermore, prolonged exposure to food-related cues has been linked to weakened cognitive con-
trol, potentially contributing to overeating and other maladaptive behaviors.
Several other studies have demonstrated that cognitive control weakens in individuals with obesity
or dieting tendencies under prolonged exposure to triggers, leading to deteriorations in eating behavior,
such as overeating (Schienle et al., 2019). Simultaneously, the anticipation of food rewards decreases
attention and increases impulsivity (Schiff et al., 2021), which is supported by improvements in cognitive
control when the reward anticipation is more distant. The automatic and controlled aspects of eating
behavior are interrelated, and avoiding contextual triggers can improve cognitive control and lead to
healthier eating behavior (Fürtjes et al., 2020). Additionally, cognitive-behavioral therapy (CBT) integrated
with cognitive training exercises has shown promise in improving inhibitory control, a key factor in prevent-
ing compulsive overeating (Manasse et al., 2020). Emerging evidence also supports the effectiveness of
combining cognitive training and CBT for regulating eating behaviors, including those within normative
ranges (Yang et al., 2019). Additionally, interventions such as guided self-help CBT have shown potential
in reducing maladaptive eating behaviors, highlighting the impact of cognitive and behavioral regulation
on managing these risks (Setsu et al., 2018).
Thus, the data on the psychophysiological mechanisms of eating behavior confirms the significance
of cognitive control and the reward system in regulating eating habits and their role in the development of
eating disorders. Investigating evoked brain activity in food preference decisions and its relationship to eat-
www.ijcrsee.com
17
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
ing behavior and general nutritional knowledge can provide valuable insights into the neural mechanisms
underlying behind rationality of food preferences and consumption. Such an approach will be contributing to
the development of more precise and personalized interventions for preventing and treating eating disorders.
Materials and Methods
The sample consisted of 40 participants aged 18 to 25 years (70% female). EEG data were col-
lected from 26 participants aged 18 to 33 years (80% female). None of the participants reported diag-
nosed disorders or complaints related to eating behavior. Participation was voluntary. All participants were
briefed on the study’s objectives and procedures and provided written informed consent.
Participants were asked to report their gender, age, and answer a series of questions regarding
general nutritional knowledge using the translated version of the General Nutrition Knowledge Question-
naire (Kliemann et al., 2016).
To study the psychological aspects of eating behavior, the following methods were employed: Dutch
Eating Behavior Questionnaire (DEBQ) (T. VanStrien et al., 1986; Russian adaptation by I.G. Malkina-
Pykh, 2007), Thoughts and Behaviors Questionnaire (TBQ) (M. Cooper, G. Todd, R. Woolrich, 2006;
Russian adaptation by A.V. Anikina, T.A. and Rebeko, 2009), Three-Factor Eating Questionnaire (TFEQ)
(A. Stunkard, S. Messic, 1985; Russian version, 2018).
To explore food preferences, a protocol for an experimental psychophysiological study was developed, which
included two types of tasks. Computerized tasks were programmed and performed using the PsychoPy software
and included high-resolution images of food items collected from the Internet (examples are shown in Table 1).
The task was modeled as a variation of the Go/NoGo paradigm. Stimuli consisted of images of
various food items. The test was conducted in two series: the first series involved selecting food stimuli
congruent with the participant’s preferences; the second series involved selecting food stimuli incongruent
with the participant’s preferences. Images of various food items that matched (first series) or did not match
(second series) the participant’s preferences were used as target stimuli with a Go response. Each series
included 150 stimuli presented in random order, with a stimulus exposure time of 500 ms.
Table 1. Examples of stimuli by series
Task Example of stimulus material
2. Go-stimuli are foods congruent with
participant preferences
2. Go-stimuli are foods incongruent with
participant preferences
www.ijcrsee.com
18
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
ERP was recorded using a “Neurovisor 136” electroencephalograph, in 128 channels (monopolar
montage with A1 and A2 ear electrodes as a references). EEG data were processed using WinEEG soft-
ware. For ERP pre-stimulus interval was limited to 200 ms, and the post-stimulus interval was set to 700 ms.
Statistical methods: distribution parameter analysis was conducted using the Shapiro-Wilk test.
Comparative analyses were performed using the Mann-Whitney U test and Student’s t-test, with effect
sizes calculated using Cohen’s d and biserial correlation coefficients. Additional analyses included one-
way analysis of variance (ANOVA), Spearman’s rank correlation coefficient, and k-means cluster analysis.
Results
To classify participants based on multiple indicators of eating behavior, a cluster analysis (K-means
clustering) was conducted. The following variables were used for clustering: restrictive, emotional, and ex-
ternal eating behavior (Dutch Eating Behavior Questionnaire); negative, positive, and permissive thoughts;
dieting and overeating; concerns about clothing, weight, and shape; food-related behaviors and the eating
process (Thoughts and Behavior Questionnaire); as well as restraint, disinhibition, and hunger susceptibil-
ity (Three-Factor Eating Questionnaire). As a result, two distinct clusters were identified, differing in the
severity of eating behavior traits, cognitive patterns, and behavioral aspects of nutrition. Cluster 1 (n = 25,
22 women, 3 men, mean age = 29) was characterized by low to moderate levels across these indicators,
reflecting a healthy eating behavior style. Cluster 2 (n = 15, 14 women, 2 men, mean age = 29) exhibited
moderate to high levels of the same indicators, suggesting a tendency toward disordered eating patterns.
Table 2. Results of the analysis of differences between clusters based on the indicators used for clustering
Indicator
Means
Between-
group
variance
Within-group
variance
F
Significance
level
Cluster 1 Cluster 2
Restrictive eating behavior 1,94 2,91 9,30 23,30 15,57 0,000
Emotional eating behavior 1,78 2,59 6,29 28,47 8,62 0,006
External eating behavior 2,76 3,18 1,68 10,69 6,13 0,018
Negative thoughts 6,46 42,05 12360,95 8351,00 57,73 0,000
Positive thoughts 4,40 30,19 6487,76 5264,44 48,06 0,000
Permitting thoughts 22,23 48,57 6769,97 10269,67 25,71 0,000
Diet 15,00 30,86 2453,85 7996,00 11,97 0,001
Overeating 14,57 38,39 5536,76 4794,88 45,03 0,000
Clothing 12,56 50,88 14322,33 10627,91 52,56 0,000
Weight and shape 7,12 39,25 10071,58 10519,64 37,34 0,000
Behavior associated with food 17,87 32,08 1971,65 12728,11 6,04 0,019
The process of eating 17,20 31,09 1883,28 8466,11 8,68 0,005
Restriction 4,40 7,75 109,49 575,00 7,43 0,010
Disinhibition 2,88 6,44 123,47 244,58 19,69 0,000
Sensitivity to hunger 3,08 4,25 13,36 172,84 3,01 0,090
In the analysis of group differences, significant variations were observed in eating behavior traits, as
measured by the Dutch Eating Behavior Questionnaire (DEBQ) (Table 2). Cluster 1 displayed scores at the
lower boundary of normative values for restrictive and external eating behaviors, with notably low scores
for emotional eating. Restrictive eating scores (1.9) suggest that respondents in this cluster do not make
intentional efforts to lose or maintain weight. Furthermore, their need for food likely aligns more closely with
actual hunger cues (external eating behavior score: 2.76). The low emotional eating score characterizes this
group as exhibiting a healthy eating style, without a tendency to use food as a coping mechanism for stress.
In contrast, Cluster 2 demonstrated elevated scores across all three scales of the DEBQ. The re-
strictive eating score (2.9) indicates a preoccupation with weight gain and attempts to restrict food intake.
Higher emotional eating scores reflect a propensity to use food as a means of coping with emotional
distress. Regarding external eating behavior, participants in this cluster are less guided by internal hunger
signals and more influenced by external food-related stimuli when making decisions about eating.
www.ijcrsee.com
19
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Eating-related thought processes also significantly differed between clusters. Respondents in Clus-
ter 1 scored low across all scales, indicating minimal engagement with negative, positive, or permissive
thoughts regarding food. Conversely, Cluster 2 exhibited substantially higher scores, mostly within aver-
age ranges, suggesting a stronger tendency to associate food with guilt, loss of control, weight gain,
and punishment. Permissive thoughts in Cluster 2 appeared to counterbalance negative ones, allowing
participants to justify food-related behaviors.
Behavioral patterns further highlighted differences: Cluster 1 scored low across all measures, while
Cluster 2 showed higher scores for behaviors such as “Clothing,” “Weight and Shape,” and “Overeating.”
These scores suggest that individuals in Cluster 2 are more prone to overeating and tend to conceal their
bodies with specific clothing to mask their shape.
The Three-Factor Eating Questionnaire (TFEQ) revealed notable differences between clusters as
well. Cluster 1 had consistently low scores across all three factors, whereas Cluster 2 showed scores for
restraint and disinhibition twice as high as those in Cluster 1, although still within the lower end of the
normative range. This suggests that participants in Cluster 2 are more inclined to restrict their food intake
to manage body weight and size and are also more prone to impulsive eating behaviors. The “Hunger
Susceptibility” factor remained within normal limits for both groups, indicating a general ability to manage
hunger and cravings in both clusters.
Considering the absence of high scores across both clusters, it can be concluded that the sample
overall does not exhibit dysfunctional eating patterns.
In summary, the two groups differ significantly in the extent of restrictive, emotional, and external
eating behaviors; the presence of negative, positive, and permissive thoughts; behavioral patterns; and
restraint and disinhibition in eating. Cluster 1 is characterized by a healthier eating style, whereas Cluster
2 shows greater concern with weight management or reduction.
Regarding differences in general awareness and knowledge about nutrition and health, Cluster 2
showed significantly higher levels of both general awareness and specific knowledge about the effects
of food on health compared to Cluster 1, which exhibited lower scores in these areas. The results of the
statistical analysis are presented in Figure 1.
Figure 1. Differences Between Clusters in General Awareness of Nutrition and Health (p<0.05)
These defferences align with the hypotheses proposed earlier. The findings likely reflect the char-
acteristics of the clusters. Since Cluster 2 is generally more concerned with weight maintenance and
reduction, as well as dietary control, their higher levels of general awareness of nutrition and health may
reflect a greater interest in these topics.
www.ijcrsee.com
20
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Figure 2. Analysis of Food Preferences Among Respondents
An analysis of average scores for food preferences among respondents in both clusters revealed
a tendency to select healthier options, particularly vegetables and fruits. Respondents from both clusters
showed the least preference for desserts (Cluster 1: 10.08; Cluster 2: 12.36). Differences in food prefer-
ences between the two clusters indicate that respondents in Cluster 2 had higher average scores across
all food categories, with a particularly noticeable gap in the selection of low-calorie foods (Cluster 1: 17.54;
Cluster 2: 22.07) and fruits and vegetables (Cluster 1: 19.71; Cluster 2: 24). Respondents in Cluster 2 also
scored higher in their preference for fatty foods (Cluster 1: 15.42; Cluster 2: 19.07).
Figure 3. Analysis of Food Choices Incogruent to Preferences
When analyzing food choices that diverged from respondents’ stated preferences, Cluster 2 partici-
pants again demonstrated higher average scores across all food categories. For both clusters, desserts
were the least chosen items. The smallest difference between clusters was observed in desserts as a cat-
egory of non-preferred food (Cluster 1: 12.13; Cluster 2: 12.43). Low-calorie foods and fruits/vegetables
were less frequently chosen as non-preferred items, with Cluster 1 participants more frequently selecting
low-calorie foods (Cluster 1: 4.17; Cluster 2: 6.14), while Cluster 2 participants favored fruits and vegeta-
bles (Cluster 1: 3.71; Cluster 2: 7.21). Fatty foods, as non-preferred items, were also selected more often
by Cluster 2 participants than Cluster 1 (Cluster 1: 6.92; Cluster 2: 9.5).
Overall, both clusters showed a general preference for healthier food options, particularly fruits
and vegetables, and the least preference for desserts. However, Cluster 2 participants displayed higher
www.ijcrsee.com
21
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
average scores across all food categories, especially in the selection of low-calorie foods and fruits/veg-
etables. These findings suggest that respondents in Cluster 2 are inclined toward more differentiated food
choices, which aligns with the overall characteristics of the cluster.
Heatmap analysis revealed higher amplitudes of evoked potentials during congruent food choice
tasks in the first series for Cluster 2 participants. These increased amplitudes were observed in early- and
mid-latency components across all types of stimuli (Figure 4).
Figure 4. Temporal Dynamics of Average Amplitude of Evoked Potentials (EP) in Food Preference Task 1 in Clus-
ters Based on Psychological Eating Behavior Characteristics (Cluster 1 - Bottom Row, Cluster 2 - Top Row)
Statistically significant differences between the clusters were found in tasks involving congruent
food choices. When accessing low-calorie dishes, differences were observed in the central-parietal region
at 100–200 ms, in the central and temporal regions at 200–300 ms, in the left frontal area at 300–400 ms,
and in the left fronto-central region at 400–500 ms. Similar differences were noted when viewing dessert
images, with distinctions appearing in the temporal region at 100–200 ms, in the central, parietal, and
frontal regions at 200–300 ms, in the frontal region at 300–400 ms, and in the left fronto-temporal and
fronto-central regions at 400–500 ms. For fruits and vegetables, significant differences emerged in the
occipital region at 100–200 ms, in the temporal regions at 200–300 ms, in the temporal region and a single
frontal electrode at 300–400 ms, and at single electrodes in the left fronto-central region at 400–500 ms.
In response to high-calorie foods, differences were detected in the central-parietal region at 100–200 ms,
in the central-frontal and parietal regions at 200–300 ms, in the left frontal area at 300–400 ms, and in the
left fronto-central and right fronto-temporal regions at 400–500 ms.
The observed temporal and topographical differences in processing various food types (low-calorie,
www.ijcrsee.com
22
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
high-calorie, desserts, fruits, and vegetables) suggest variability in cognitive and emotional responses de-
pending on the food category. More pronounced differences in the temporal and frontal regions for desserts
and high-calorie foods may indicate greater involvement of reward-related and cognitive control areas.
Figure 5. Analysis of Differences in Average EP Amplitude in Food Preference Task 1 in Clusters Based on Psy-
chological Eating Behavior Characteristics (p<0.05)
When performing the task of making food choices incongruent with dietary preferences in the sec-
ond series, a higher amplitude of evoked potentials (EP) of early and middle latency was also observed
in the second cluster, regardless of stimulus type (Figure 6).
www.ijcrsee.com
23
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Figure 6. Temporal Dynamics of Average Amplitude of Evoked Potentials (EP) in Food Preference Task 2 in Clus-
ters Based on Psychological Eating Behavior Characteristics (Cluster 1 - Bottom Row, Cluster 2 - Top Row)
Statistically significant differences between clusters were identified during the task of making food
choices incongruent with individual preferences for specific stimulus types. High-calorie food images elic-
ited differences at 100–200 ms in the central-parietal and left temporal regions. The 200–300 ms interval
showed differences in the frontal-central, left temporal, and temporo-parietal regions. Between 300 and
400 ms, significant activity appeared in the left frontal and central regions, while the 400–500 ms interval
involved the left frontal-temporal region and a single right frontal-temporal channel. For dessert images,
no differences were observed at 100–200 ms; however, significant activity emerged during the 200–300
ms interval in the central, left-lateralized frontal, and parietal regions, followed by activity in the frontal
region at 300–400 ms and the left frontal-central and temporal regions at 400–500 ms. Low-calorie food
images showed differences during the 100–200 ms interval in the central-parietal region. At 200–300 ms,
significant activity was detected in the left-lateralized frontal, central, and temporal regions. By 300–400
ms, activity persisted in the frontal and central regions, with left lateralization, and at 400–500 ms, in the
left frontal-temporal region. For fruits and vegetables, differences were first noted during the 100–200 ms
interval in the central-parietal region. Subsequent differences were observed at 200–300 ms in the left-
lateralized frontal, central, and temporal regions, followed by activity at 300–400 ms in the left-lateralized
frontal and central regions, and at 400–500 ms in the left frontal-temporal region.
www.ijcrsee.com
24
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Figure 7. Analysis of differences in averaged EP amplitude values in Food Preference Task 2 in Clusters Based
on Psychological Eating Behavior Characteristics (p<0,05)
Overall, significant differences between clusters identified by eating behavior traits were predomi-
nantly observed in the temporo-occipital regions across all stimulus types. For high-calorie food, Cluster 1
showed higher amplitudes during early processing stages (100–200 ms), while Cluster 2 exhibited greater
amplitudes at later intervals (200–300 ms and 400–500 ms). For low-calorie food and fruits/vegetables,
significant differences were primarily localized in the parietal-occipital regions, with Cluster 1 demonstrat-
ing consistently higher amplitudes in early intervals (100–300 ms). These findings suggest distinct neural
processing patterns between clusters, reflecting differences in attentional or evaluative responses to food
stimuli. These findings highlight the role of cognitive control and attentional modulation in decisions that
deviate from habitual food preferences, reflecting their importance in regulating eating behavior.
Next, to test the hypothesis that brain activity characteristics in food preference choice tasks may
differ depending on the level of awareness in nutrition and health, a comparative analysis was conducted
between the group with above-average results (group 1) and the group with below-average results (group
2). Respondents with average results were excluded from the sample during the calculations.
www.ijcrsee.com
25
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Visual differences between the heatmaps of the groups with different levels of awareness in nutri-
tion and health are minimal. However, an increase in the amplitude of event-related potentials (ERPs) was
observed in the first group of participants, characterized by higher levels of awareness, in the occipital
area between 100-400 ms during the first series, regardless of the type of stimulus (Figure 8).
Figure 8. Temporal Dynamics of Average Amplitude of Evoked Potentials (EP) in Food Preference Task 1 in
groups depending on the level of awareness in nutrition and health (results for group 1 in the lower row of im-
ages, for group 2 in the upper row).
Since the distribution in the selected groups for several indicators differed from normal (Shapiro-
Wilk test), the analysis was conducted using both a parametric criterion (Student’s t-test) and a non-
parametric criterion (Mann-Whitney U-test). Significant differences between the first and second groups
in the congruent food choice task during the first series, when demonstrating images of high-calorie foods,
were found in the 100-200 ms interval in the temporal areas, in the 200-300 ms interval in the left and right
temporal areas, and in the 400-500 ms interval in the left central-temporal area. No statistically significant
differences were found in the 300-400 ms interval (Figure 9).
www.ijcrsee.com
26
Ermakov, P. et al. (2025). Evoked Brain Activity in Food Preference Decisions: Links to Eating Behavior and General Nutritional
Knowledge, International Journal of Cognitive Research in Science, Engineering and Education (IJCRSEE), 13(1), 15-31.
Figure 9. Analysis of Differences in Average EP Amplitude in Food Preference Task 1 in groups depending on the
level of awareness in nutrition and health (p < 0.05).
Statistically significant differences in averaged ERP amplitude values across participant groups
depending on the level of awareness in nutrition and health were predominantly localized in the temporo-
occipital regions. For dessert images at 100–200 ms, differences emerged in the parietal-occipital regions,
in the 200–300 ms interval, significant activity was detected at a single electrode in the fronto-temporal
and temporal regions, while at 400–500 ms, differences were observed in the left temporo-parietal, fronto-
temporal, and right frontal regions. For low-calorie dishes, significant differences appeared at 100–200 ms
in the right temporo-parietal region. During the 200–300 ms interval, differences were noted at individual
electrodes in the central, parietal, and left parieto-occipital regions. By 400–500 ms, significant differ-
ences were observed at individual electrodes in the temporal and fronto-central regions. When fruits and
vegetables were presented, differences were observed at 100–200 ms at individual electrodes in the left
parieto-occipital, fronto-temporal, and right parieto-occipital regions. At 200–300 ms, significant activity
was seen in the left centro-temporal and right parieto-occipital regions. Finally, during the 400–500 ms
interval, differences were observed at a single electrode in the right temporal region.