The triple network model is a way of understanding how the brain “decides” if it should pay attention to the inner world or outer world (Menon 2011). The three networks are:
Frontal-Parietal network (FPN) (also known as the Central Executive Network, CEN)
Default Mode network (DMN)
Salience network (SN)
With this model, Menon proposed that the SN acts as a filter to all incoming stimuli, external to the brain (sensory or limbic) as well as internal (like memories) which come from the DMN. If the stimuli are not deemed important, the DMN is activated, but if it is, the FPN is activated which empowers the attention network. The FPN can enact top-down control over the SN to maintain focus on a task towards a certain goal.
This model is widely applied within neuroscience research, but it is worth noting that it was originally proposed as a framework to understand mental illnesses (Menon 2011).
“Deficits in access, engagement and disengagement of large-scale neurocognitive networks are shown to play a prominent role in several disorders including schizophrenia, depression, anxiety, dementia, and autism.” (Menon 2011)
The inclusion of the attention network, made up of the dorsal and ventral attention networks (DAN and VAN), adds a level of detail regarding attention switching control in the brain. Su et al studied the impact on activation and connectivity between four networks (Su et al. 2021).
Frontal-Parietal network (FPN)
Default mode network (DMN)
Salience network (SN)
and Attention Network (DAN & VAN)
The attention network balances two regions of the brain (Su et al. 2021). The dorsal attention network, which is involved in top-down processing and the ventral attention network, which does bottom-up processing (Su et al. 2021). So, the DAN-VAN connection balances voluntary and involuntary attention.
By the connectivity they measured, a modified version of the Triple Network model is observed. The SN still filters information and sends signals to the DMN or FPN (CEN), but in this model, the FPN also regulates the DMN and engages the attention network for cognitive control of attention.
Research by Su et. al in 2021 identified several impacts on these network interactions while participants watched either personalized or generalized video clips via the TikTok app. Their results are summarized in “Figure 6” below (Su et al. 2021).
On the left is the DMN, made up of the midline core, medial temporal lobe (MTL), and the dMPFC. In the center are the FPN (or CEN) and the SN. On the right is the attention network, DAN and VAN.
Mapping these connections to the above simplified network model, we see strengthened coupling within the attention network (DAN-VAN) and stronger coupling between the FPN and the AN, as well as the FPN-DMN.
⬆️DAN-VAN The brain is both voluntarily and involuntarily interested in the videos. This makes sense because the participant was actively watching the videos, an external stimulus.
⬆️FPN-DAN Increased coupling between the FPN and the DAN is the brain telling itself to continue paying attention to external stimuli (i.e. the videos).
⬆️FPN-DMN It has been theorized that this connection indicates that the FPN mitigates internal processing in the DMN, in favour of processing the external stimuli.
The triple network model above has been highly simplified in order to facilitate generalized understandings. There are more connections between these networks that are important to note as they are activated when watching TikTok.
⬆️VAN-DMN Increased coupling between the VAN and the DMN (specifically the dMPFC, involved in thinking about oneself) suggests the viewer’s brain is processing how the new information relates to themselves. The VAN recognizes this as important since the stimuli seem to be for them.
NOTE: This coupling was observed only when the TikToks were personalized and was not observed for general content.
⬇️VAN-SN Down-regulating the connection between the SN and the VAN is concerning. Remember from the triple network model that the SN is the gatekeeper of incoming stimuli, deciding if it is worth the attention of the brain. The result of decreasing this coupling is that attention shifts more by reflex than by learned salience.
The brain becomes susceptible to jumps in attention in response to things like:
Flashing lights
Sudden movements
Abrupt changes in volume
Short form media capitalizes on these instinctual attention-grabbers to keep us hooked.
Interestingly, the SN contains both the amygdala and the ventral tegmental area (VTA) which are both activated when watching short form media (Gao & Hu, 2025). To know that the VAN-SN connection is simultaneously down-regulated seems counter-intuitive and was indeed counter to their hypothesis (Su et al. 2021).
A final note on brain connectivity is within the DMN, where reduced coupling of the midline core with the MTL is observed (Su et al. 2021).
⬇️Core-MTL Decreased coupling of these regions within the DMN results in the decreased occurrence of spontaneous thought, like daydreaming, and generation of our inner monologue.
Scrolling reels or sitting on TikTok doesn’t just pass time, it reconfigures how your brain talks to itself. By studying functional connectivity of the brain during scrolling, we see our attention is monopolized. Algorithms trick the brain into assigning false importance to content by mimicking personal significance. We are drawn in over and over, as our instincts direct our attention to each new video. Our stream of consciousness is being over-written by external noise and our daydreaming put on hold. Spontaneous, creative thoughts, the ideas that make us human, are being suffocated by the machine that is the feed.
References
Menon, Vinod. “Phenomena of the human mind which are taken advantage of by shortform media: creating frameworks to understand, seeing patterns, the comfort of familiarity BRANCH-OFF Summary of exploited human biases How to improve your open-mindedness.” Trends in Cognitive Sciences, vol. 15, no. 10, 2011, pp. 483-506, https://www.sciencedirect.com/science/article/abs/pii/S1364661311001719.
Su, Conghui, et al. “Individualized video recommendation modulates functional connectivity between large scale networks.” Hum Brain Mapp, vol. 42, no. 16, 2021, pp. 5288-5299, https://pubmed.ncbi.nlm.nih.gov/34363282/.