1. Introduction
Problematic use of social media (SM) platforms is an ever-growing concern that affects various populations, especially people from low-income countries [
1], with moderate to low school achievement, low parental control [
2], low self-esteem [
3], feelings of loneliness, fear of negative evaluation, and behavioral inhibition [
4]. Adolescents seem to be amongst the more vulnerable populations due to SMs’ unique appeal to their age range. They are drawn to these platforms for social interactions [
5]. However, social feedback plays a critical role in shaping social behavior and well-being throughout life, from childhood to older adulthood. In early development, positive social feedback from peers and caregivers helps foster social skills, self-esteem, and a sense of belonging. As individuals age, social feedback continues to influence behavior, reinforcing social bonds and maintaining relationships. In older adults, social feedback becomes increasingly important due to the reduction in social networks, leading to potential feelings of loneliness and isolation (e.g., Tragantzopoulou & Giannouli, [
6]). Positive social interactions and feedback can mitigate these effects by promoting emotional well-being and reducing feelings of loneliness, while the absence or negative social feedback may exacerbate loneliness and contribute to mental health decline in this population.
SM dependence is defined as the excessive and uncontrolled usage of SMs, leading to impairments in personal, social, and professional aspects [
7,
8]. Excessive users of SM can show abstinence symptoms, such as anxiety [
9,
10], stress, and depression [
9,
11]. They can also develop a tolerance, resulting in the need to progressively increase the time spent on SMs to obtain the same level of gratification experienced in the early stages. Another relevant aspect of dependence is the excessive worry associated with SM-related behaviors, and a lack of self-control with internet usage [
7]. However, other results show no negative outcomes, only showcasing some specific behaviors as potentially dangerous [
12]. One of the reasons for this discrepancy is that SM dependence could also be explained based on the person’s preferred SM, as some results show that changing behaviors could be more difficult in certain SM [
13] Given these differences, this type of addiction is still being debated in the literature, with the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; [
14]) and International Classification Diseases (ICD-11; [
15]) not recognizing addiction to SM as a separate diagnostic category, despite it being studied as a behavioral addiction when it becomes excessive and affects the daily life of an individual [
8,
16].
In the context of SM, the “like” button acts as a reward mechanism, providing quick and simple feedback on users’ social media activity [
17]. Given its purpose, the “like” feature can motivate users to adjust their publications and online behaviors to maximize the chances of receiving this reward [
18].
The Interaction of Person-Affect-Cognition-Execution (I-PACE) model [
19] offers insights into the underlying mechanisms involved in the development of various behavioral addictions. The I-PACE model highlights the importance of the interaction between predisposing factors, affective and cognitive responses to stimuli, and executive functions—such as inhibitory control and decision making—in the development of behavioral addiction. In the case of social media, like gambling or gaming disorders, the process may involve heightened cue-reactivity and craving, coupled with diminished inhibitory control, which can contribute to habitual behaviors. Further studies are needed to explore both the common and distinct mechanisms involved in addiction, obsessive–compulsive disorders, impulse control disorders, and substance use disorders, as these conditions share underlying neurobiological processes.
Studies have used functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and evoked-related potentials (ERP) to better understand the effects of social feedback through the “like” on brain activity. Each technique offers advantages and limitations. fMRI has excellent spatial resolution, making it possible to identify the brain regions involved in the perception and processing of “likes” [
20]. On the other hand, EEG, due to its high temporal resolution, is especially suitable for studying rapid cognitive processes such as attention and sensory responses.
fMRI studies show that the amygdala, ventromedial prefrontal cortex (vmPFC), and ventral striatum are key structures in reward processing [
21,
22,
23]. The vmPFC plays a key role in this network through the observation and evaluation of the reward (i.e., “number of likes”; followers), motivating behavior towards it. The amygdala further solidifies the behavior-oriented toward the positive outcome, which can lead to SM addiction. Additionally, using SM may be associated with reduced amygdala volume [
22]. The striatum, which shares various connections with different brain structures, allows the integration of information from various modalities and is necessary for learning and evaluating rewards [
21,
23], which in turn supports motivated behavior [
23]. The involvement of the striatum in reward processing extends to social situations, where certain situations, such as a compliment from a colleague, result in increased activity [
21].
For EEG data, evoked potentials, specifically the P300 and N200 components, have been used to investigate online social interactions. The P300 is associated with the allocation of attention and processing of relevant stimuli [
24,
25,
26]. In the context of social networks, an increase in the amplitude of the P300 suggests a greater allocation of neuronal resources to the processing of these socially relevant stimuli. The N200 is often associated with conflict detection and decision making [
27].
Current research shows the activation of several structures related to reward processing and the impact that their activation has on continued engagement with SM. As such, the need for research that further validates the effects of certain aspects, such as the “like” feature, on continued engagement is increasingly present. Despite some online social interactions having similarities with offline interactions, they are not identical. For instance, reward processing differs, as online social rewards lead to higher activation of structures such as the nucleus accumbens (NACC) while showing less amygdala recruitment [
22]. Identifying these differences, we aim to systematically review studies on the neural correlates of online social feedback, using EEG and fMRI as investigative techniques. The main aim is to systematize the neuronal correlates and neurobiological processes that underlie the response to social feedback, focusing on how this stimulus is processed by the brain structures involved in social reward processing. Considering the relevance of investigating the “like” feature of social media platforms through fMRI and EEG (e.g., Meshi et al. [
28]; Bhanji & Delgado [
21]), the following are the proposed main research questions:
Research question 1: What are the effects of the “like” feedback on neuronal regions associated with reward processing, assessed through fMRI?
Research question 2: What are the effects of the “like” feedback on brain activity, assessed through EEG?
Additional questions involve exploring potential differences in brain activity associated with feedback valence (i.e., positive versus negative) and comparing habitual versus sporadic social media users. Behavioral studies on this topic will also be reviewed.
4. Discussion
This systematic review aimed to analyze neurophysiological studies (i.e., studies using fMRI and EEG methodologies) investigating online social reward processing, mainly the feedback provided through the “like” feature.
The review found significant variability in the objectives, methods, and results across the studies. The objectives often align with key social factors influencing social media (SM) use, such as social rewards [
13,
43], usually in the form of a “like”, reward learning, and decision making [
23].
Regarding the main research questions—what are the effects of the “like” feedback on neuronal regions associated with reward processing, assessed through fMRI and through EEG?—the analysis of the data related to receiving feedback, both positive and negative, revealed activation in several brain structures associated with reward processing, including the striatum and thalamus [
28,
39]. The structures involved in goal-oriented and social behavior, such as the ventrolateral prefrontal cortex (vlPFC) and medial prefrontal cortex (mPFC) [
44,
45], are also engaged in feedback processing. These findings align with the existing literature that underscores the role of these structures in social reward processing [
21,
22,
46,
47].
Differential brain activity for positive versus negative feedback was also identified. The nucleus accumbens (NACC) appears particularly crucial for engagement with SM, as its activation correlates with positive feedback and the intensity of SM use [
28]. This supports the notion that the NACC plays a significant role in motivating behavior related to positive social gains and avoiding social punishment [
48]. The insula is involved in reward anticipation and avoiding social punishment [
46], a finding corroborated by the reviewed data [
41].
The amygdala’s activation in response to positive feedback is expected [
35], given its role in social behavior coordination [
49] and positive outcome processing [
22]. In contrast, the results for negative feedback were less extensive, primarily showing activation in the vlPFC and mPFC [
41]. This gap highlights the need for future research into the neuroanatomical areas involved in processing negative feedback. Broader research on social rewards has identified the NACC, insula, and right inferior frontal gyrus as sensitive to negative social outcomes [
46,
48,
50].
Specifically, when it comes to receiving a “like”, the striatum, thalamus, hippocampus, and VTA are activated [
17]. This is consistent with their roles in various aspects of social rewards and behavior [
21,
23,
46,
51]. Notably, the insula, amygdala, ventromedial prefrontal cortex (vmPFC), and paracingulate cortex are uniquely activated in response to receiving a “like” rather than giving feedback [
18]. However, it is worth noting that some expected structures, such as the NACC, which is typically involved in processing positive feedback [
28], were not identified in this study [
17].
The significance of the “like” feature as social feedback is further supported by studies examining the effects of the quantity of “likes” on brain activity. Similar activation patterns were observed in structures such as the mPFC, NACC, and thalamus [
36]. Additional structures activated by receiving many “likes” include the caudate, brain stem, VTA, and hippocampus [
36]. The VTA and hippocampus are associated with memory and pro-social behavior [
47,
51,
52] and contextual memory for social rewards [
53], respectively. The caudate and brain stem are also integral to models of social reward processing [
46], with the caudate being involved in both monetary and social rewards [
54]. Despite their roles in social behavior, these regions were not consistently reported, with some appearing only in specific paradigms related to feedback valence or quantity of positive feedback (i.e., number of “likes”). Understanding these discrepancies is crucial for elucidating the impact of social feedback on SM.
Furthermore, EEG data revealed predominantly beta wave activity in response to receiving many “likes” [
38]. Given beta waves’ involvement in decision making and selective attention [
55], EEG findings corroborate fMRI data, highlighting the activating effect of numerous “likes”.
Overall, the results show considerable variability in effects, and while not contradictory, they do not establish a consistent activation system for online social rewards. Methodological differences and various influencing factors, such as the characteristics of the feedback giver and recipient, including sex and personality traits, can affect reactions to positive feedback. For example, a more positive rating from the feedback giver was associated with the amygdala and vmPFC [
35]. This suggests that higher social hierarchy may involve greater amygdala activation due to its role in hierarchical learning and reward processing [
56]. The vmPFC’s role in processing social feedback and self-relevant information [
57] might explain its activation when receiving feedback from individuals with higher ratings. Additional structures, such as the hippocampus and mPFC, also relate to social hierarchy [
56], indicating a need for further research into their involvement in feedback processing from high-status individuals. While evidence on sex differences in processing social feedback is limited, one study indicated increased activity in the right OFC and right anterior insula in response to feedback from the opposite sex [
35]. The particular activation of these structures could be related to somatic activation [
58] or sexually relevant stimuli, such as attraction toward the opposite sex [
59]. Personality traits may also influence SM engagement through rewards tailored to specific traits, affecting brain activity such as the P3 component [
25].
Non-neuroimaging studies reinforce the importance of the characteristics of the giver and recipient of the “like”. On one hand, the characteristics of the giver are shown to be more relevant in determining the subjective value of the “like” than the number of “likes” received [
60]. On the other, the characteristics of the receiver (i.e., age) seem to alter the perceived value of the “like” [
61]. Personality is one of these characteristics, with certain personality profiles, such as narcissism [
62], leading to higher SM use [
63,
64], further modulated by sex and age [
65]. Indeed, these individual characteristics seem to be potential vulnerabilities towards a pathological engagement with SM, as several studies have shown that variables such as impulsivity, narcissism, fear of missing out, and low self-esteem can be associated with an increased risk of developing this kind of involvement [
66]. As such, and as pointed out in some studies [
35] the neural correlates of social feedback could vary based on the characteristics of the giver and recipient.
The findings from our systematic review offer valuable insights into the neural correlates of social media feedback, particularly the “like” feature, and its impact on brain activity.
One of the key clinical insights is the role of brain structures such as the nucleus accumbens (NACC), vmPFC, and amygdala in reward processing. The activation of these regions in response to positive social feedback highlights the neurobiological underpinnings of social media use, which can inform clinical approaches to addressing problematic social media behavior, particularly in adolescents. The excessive activation of these reward-related brain areas, especially the NACC, is linked to increased intensity of social media use, which may contribute to the development of dependency or addiction-like behaviors [
28]. This suggests that interventions aimed at reducing social media use could benefit from targeting the reward systems and enhancing self-regulation strategies to counterbalance excessive reward-seeking behaviors.
The review highlights the amygdala’s role in social behavior and its involvement in processing positive feedback. Given that the amygdala is also associated with emotional regulation, its activation in the context of social media may explain why some individuals are more prone to anxiety or mood disturbances in response to online interactions [
22]. Clinically, this underscores the need for mental health interventions that address the emotional impact of online social interactions, particularly in populations vulnerable to anxiety or depression.
Additionally, the findings suggest a significant difference in how positive and negative feedback is processed in the brain. The relatively limited involvement of brain structures in response to negative feedback, particularly when compared to the robust activation seen with positive feedback, may indicate that users of social media, particularly adolescents, are more sensitive to social rewards than to social punishments. Clinically, this raises concerns about how the continuous pursuit of social validation may reinforce maladaptive behaviors, such as excessive engagement with social media, while potentially diminishing resilience to social rejection or negative feedback [
41].
There are similarities between the brain structures involved in processing the “like” feedback and the structures involved in processing gambling outcomes. For example, the activation of the amygdala and the ventromedial prefrontal cortex (vmPFC) to gambling outcomes could be an indicator of problematic gambling [
67,
68]. Non-neuroimaging data corroborates the similarities between conditions, as evidenced by increased decay of the reward value over time (delay discounting) [
69,
70], or by problematic engagement possibly serving as a coping mechanism with adverse consequences [
19,
71]. Furthermore, individuals with more severe SM-related problematic behavior are more likely to exhibit additional problematic behaviors or even develop a behavioral addiction, such as gambling disorder [
72]. Despite the similarities, a “like” functions as a social reward, in contrast to the monetary reward associated with gambling outcomes. One key difference lies in the value of a “like”, which is influenced by the characteristics of the giver. However, to our knowledge, no studies have directly compared these two types of rewards to clarify their differences. In light of these findings, future research should explore the longitudinal effects of social media engagement on mental health, especially in individuals with pre-existing vulnerabilities such as low self-esteem or social anxiety. Understanding how these neural patterns evolve could inform prevention strategies for adolescents and young adults at risk of developing dependency on social media platforms.
Clinically, the evidence from our review calls for the development of therapeutic interventions aimed at moderating the neural reward systems activated by social feedback. Cognitive–behavioral interventions focusing on self-regulation, emotional resilience, and the management of online interactions could help mitigate the potential negative impacts on mental health. Additionally, educational programs for adolescents, focusing on the psychological mechanisms behind social media usage, may empower them to navigate online spaces more consciously and healthily.
Despite the contribution of this study to the literature, some limitations must be considered when interpreting the results. One of the main limitations of this systematic review is the heterogeneity of methodologies across the included studies. The included studies used different neuroimaging techniques, including fMRI and EEG, each of which has distinct strengths and limitations. For instance, fMRI provides excellent spatial resolution, allowing the identification of specific brain regions involved in reward processing, but it lacks the temporal resolution needed to capture the fast dynamics of brain activity. On the other hand, EEG offers high temporal resolution but lacks spatial specificity, making it difficult to identify the exact brain regions responsible for certain neural responses. This variation in neuroimaging methods hinders the direct comparison of findings across studies and introduces potential biases related to the limitations inherent to each technique.
Additionally, there was significant variability in the participant demographics across the studies, particularly regarding age, sex, and social media usage habits. Most of the studies included adolescents and young adults, but the age ranges varied, and some studies did not report detailed demographic data, such as sex distribution. This heterogeneity can influence the results, as neural responses to social feedback may differ by demographic differences, as well as personality traits or susceptibility to social influence. As a result, it is difficult to generalize the findings to a broader population or to make definitive conclusions about how different groups may respond to social media feedback. Also, the studies had relatively small sample sizes, which may affect the statistical power of their findings. These limitations make it difficult to synthesize the results and draw comprehensive conclusions. Overcoming these limitations in future research will be crucial for gaining a more complete understanding of how social media feedback influences brain activity.
5. Conclusions
The current findings on the neural correlates of online social feedback, focusing on the effects of the “like” feedback on brain activity, using EEG and fMRI, show a fragmented picture, with various findings that, while not contradictory, do not clarify a distinct network of brain structures involved in processing feedback, such as the “like” button. To better understand this, it is crucial to consider a wider range of influencing variables, including sex and personality traits, which have been shown to affect brain activity. Additionally, peripheral physiological indicators could offer valuable insights into users’ reactions to SM and their subsequent behaviors.
It is important to note that giving or receiving a “like” involves different brain processes, and our review focused solely on the experience of receiving a “like”. This limitation means that our findings might not fully capture the nuances observed in studies examining both giving and receiving “likes”. Another limitation arises from some studies using alternative forms of positive feedback instead of the standard “like” button, which makes difficult the systematic analysis of results.
To enhance the validity of current findings, the increased use of EEG, due to its superior temporal resolution, would be beneficial. Comparing excessive SM users with regular or healthy users could also provide more precise insights into group differences.
In summary, as SM becomes increasingly integrated into daily life, more research is needed to explore the individual and SM characteristics, namely the role of the “like” feedback that drives user engagement. It is crucial to examine a wide range of variables to ensure that research remains relevant in the rapidly evolving field of social media, and the “like” feedback should be among them. Robust evidence is necessary to support and refine the theoretical models of online social rewards and to develop effective prevention and intervention strategies.
