top of page

Short Videos | Breakdown of the Mechanism, Hooks, and Addiction

Updated: Jun 7, 2022

When was the last time you scrolled through Instagram Reels, stories, YouTube shorts, or TikTok and realized that you have spent way more time than you intended to? I would guess it was not so long ago.

The short video format has become extremely prevalent in the last years simply because platforms know it keeps us hooked and engaged with content for longer periods of time. In this article, we'll be delving into the effects of this behavior on the dopamine system, and the underlying mechanism that eats our time raw and is irresistible, and addictive.

NB - the first two sections start with common knowledge for the people in the field, so you could skip straight to the next segments :)


With a flood of new videos that last only a couple of seconds, and the ability to instantly transition to a new video with a swipe, the brain simply does not have enough time to fully scan the environment, decide that there is nothing interesting in the content, and get bored. Furthermore, with every new video, the expectation of the next one being entertaining keeps our dopamine levels up and keeps us swiping.

This effect (of no second to waste/don't give the user time to get bored) is most prominent on TikTok where from the moment of opening the app a video starts playing straight away. This video is acting as a cue to trigger scrolling behavior which releases spikes in dopamine and creates the famous cue-trigger-action-reward loop of addictive behavior. For this reason, the first video is the one you're most likely to love (according to the algorithm), and opening the app becomes associated with this positive dopamine surge. For this reason, our brain remembers that opening the app is related to a positive dopamine response and a person is more likely to click on the app icon.

With an evolutionary novelty-seeking trait, an underlying physiological mechanism has developed (or vice versa). Dopamine is well known to be the main factor when it comes to addictive behaviors. Novelty seeking is no different in that sense.

A 2014 study published in Behavioral Neuroscience tested this hypothesis on monkeys. According to the study, novel stimuli excite dopamine neurons, and vice versa, dopamine promotes novelty-seeking behavior. The study also found that excessive novelty seeking (characteristic of behavioral addictions) might be caused by increasing dopamine levels (through lower dopamine reuptake). What this means for short videos is that we are programmed for novel stimuli to excite our brain, and this creates a behaviorally addictive mechanism, where more scrolling through short videos provides more and more novel content which translates into a constant rush of dopamine. Moreover, this system utilizes another aspect of dopamine, something often referred to as the "slot machine effect."


Slot Machine effect is closely tied with the mechanisms that make novelty rewarding. One of the key aspects of this behavior is that after a certain degree of repetitions, a habit is formed, a person who is addicted releases almost the same amount of dopamine when losing as winning because the true reward of the behavior is the expectation of the reward when pulling the lever (or swiping) rather than the reward itself.

When it comes to dopamine, people are often confused, many believe that it is a reward-associated response, while it is in sense true, in reality, it is more clearly correlated with the expectation of the reward, rather than the reward itself. For this reason, the highest level of dopamine is released right before the expectation of the reward is the highest and not when the reward is actually received. This is why when someone who is addicted to a certain behavior (and in the vast majority of gambling cases has an expectation of innate ability in the random game of chance) enjoys action that is destructive, but at the same time pleasurable. Additionally, the frequent spikes in dopamine affect the dopamine benchmark (average levels of dopamine, which circulates in the body), which makes it harder to be satisfied by other activities which otherwise could have been pleasurable, and further reinforce the behavioral loop (discussed in more detail in later sections).

Not knowing whether or not the next video is going to be rewarding in a sense of addiction is worse, because not knowing creates the expectation of the reward which hijacks our brain and releases a similar level of dopamine/joy as an entertaining video you saw that day. If the next videos were laid out at the bottom of the screen there would be a higher probability that you would look at them see that they are not rewarding and leave the app, and this is why on Tiktok, Stories, Reels, Youtube shorts the mechanics of previewing are eliminated. This is a part of the reason why now more people spend time in those sections rather than feed, and other sections where you have a choice of what would be the next content to consume. This mechanism eliminates the choice to decide what to consume and leaves it fully to the algorithm which can be quite unhealthy, and a matter for a separate discussion. Also, if someone could explain what the inscription of "Georgia" on the vending machine in the picture means, I would be infinitely grateful.


The ways we navigate different social media platforms and general UI (User Interface) vary, but when it comes to the newer platforms (TikTok) and extensions (Reels) one clear change is noticeable and it is not random. When we navigate to these videos, Primary Action Behaviors (PAB) are maximized, and secondary actions are preferred to be minimized. Swiping up/down has become the PAB on the short video platform to a degree that in order to navigate to different pages other swiping motions have either not been integrated or have been left out. For example, if you open Instagram, swiping left or right allows you to leave the feed, and go to messages or uploads, but once you enter the Reel world, this action does not work. There are a couple of reasons for these changes and why they work, some include:

  • When a user enters the part of the platform with the highest time spent after the initial visit, the objective of the platform is to maximize their stay in this section, so making leaving a section a little bit more inconvenient can have a massive result for retention of the user that is engaged in habitual behavior since even small inconveniences can prevent them from breaking the habit loop.

  • When inciting habits or behavioral patterns, the bedrock for formation is training and repetition. By maximizing behavior on the primary action which is swiping down (rather than tapping +swiping, like in stories or any other combination of actions) for the same amount of time spent on the platform a user is maximally trained to engage in this primary behavior and form a behavioral habit much faster and to a stronger degree.

This is why even tiny changes in the UI are very important and each change likely has a behavioral reason. When it comes to short video modes, the UI is highly optimized to train for desired behavior and keep users on the desired page for longer periods of time.

NB: Other interesting navigation-related changes, now Instagram Reels Showcase is visible every time someone scrolls through their feed because it is another opportunity to move users to a higher retention Reels page. The "+" button was also removed from the bottom center bar of Instagram's feed and was replaced with a Reels button. Reels also have a comparatively larger visual section in the search tab, so you are more likely to move to the Reels page.


It comes as no surprise that these short video algorithms and algorithms of social media platforms, in general, are becoming exceptionally good at predicting and suggesting content which we would engage with. The difference between the general feed on Facebook and Instagram vs Reels and TikTok is that for general feeds algorithms have to work (to a large degree) with datasets from pages and people we follow, but when it comes to short video cousins, the dataset consists of all the aggregated content around the world. This means, that the level of engagement/interest-based personalization can be much more precise, and the algorithm might know our interest by matching it with similar accounts even before we realize that the suggested subject might be interesting to us. When it comes to products and marketing, implications can be huge.

While on social media we generally (excluding ads) sift through products and services that are already in our consideration set, short video platforms can seamlessly integrate products that are outside of it into the feed, without an out-of-place feeling of an ad, for example, in the Facebook feed. In addition, a larger and more heterogeneous data pool even if we assume the same amount of collected data points gives these platforms more precision when it comes to targeting, and matching users with desired products. They can seamlessly test subjects that our online behavior might not indicate we would be interested in, and if we show interest, create a new data point. Moreover, as many interested parties have found, TikTok collects more data points as well (this Wired article might be interesting) and will eventually be an insanely powerful platform for advertisers (assuming that it will catch on with older demographic as the time passes on, which considering general adoption curve for Facebook, Instagram, etc would not be surprising).

However, the big question of interest is - Does this feed of super-personalized video content affect our brains differently, and if yes, how?

If you would like to skip some neuroscience jargon, the short answer is - very likely yes, and negatively but the paragraphs below explain in more detail.

Personalization of content is a double-edged sword, on the one hand, everyone wants to see videos that are relevant to their interests, but on the other hand, elimination of the choice element leads to a passive brain activity which can be detrimental to brain health/activity. A recent study published in the NeuroImage (2021) examined FMRI scans of users and conducted addiction tests; The methodology that was used in the study has been widely used and applied to different platforms and so far TikTok scored the highest in terms of undesired and addictive behaviors of the users.

Brain activation for two types - Personalized Videos (Pvs) and Generalized Videos (GVs) was tested with FMRIs (GVs were TikTok videos for new users before the algorithm had enough input to start personalizing them).

The test found that the primary activation areas while using the app include Default Mode Network (DMN) and Ventral Tegmental Area. These areas are usually associated with rest states when a person is not involved with the world and regulating reward behavior. However, a higher level of PCC (dorsal posterior cingulate cortex) activation and a lower level of activation for Default Mode Network (DMN) was observed for personalized videos rather than GV. PCC is part of the overall DMN, while DMN is related to overall introspective states, as the clear role of PCC in the DMN is still debatable, this result can't really tell us if a user is more "mindless" while scrolling personalized or generalized content, but regardless, a high activation of DNM was observed. As for the regions responsible for reward learning and addiction (ventral tegmental area (VTA), substantia nigra (SN), nucleus accumbens (NA)), although both types of content achieved significant activation in the SN region, activation of NA was not significant in either case and only personalized videos achieve a significant activation in VTA. VTA is generally activated when a reward is received and is largely responsible for reward, and motivation, it has been argued that this activation is necessary for developing addictions (Kalivas. 1993). Thus it is possible that the high level of personalization of the videos that are already processed in DMN further adds an addictive element to these platforms.


We've all probably heard a saying - Better an egg today than a hen tomorrow Well, it appears that the saying illustrated the brain's priority encoding pretty precisely. A study published in Proceedings of the National Academy of Sciences (Enomoto, 2011) illustrates this as well. Brain activity in monkeys doing multiple tasks with different rewards was measured and it showed that monkeys valued future rewards and included them in decision making (future reward was encoded in dopamine neurons). However, just as the saying suggests, future rewards were discounted to the present value. This means that the default brain mode, at least in monkeys, makes them prone to decisions that value instant rewards more than similar or possibly better rewards in the distant future.

If we expand this behavioral tendency to interactions on social media, we can theorize that the addictive mechanism providing instant dopamine release (of large amounts) has an unfair advantage over activities and behaviors that would also be useful and provide a similar amount of dopamine in the future. This would further add to the lack of desire to act toward activities that could possibly be beneficial in a long term.


To sum it up, short-format videos are optimized to capture our attention, release significant amounts of dopamine and create a behavioral loop. When more triggers become associated with engagement habits, those habits become more and more engrained. The overall result of these mechanisms is that we form habits and these habits make it much more likely that we go back to a default action/thinking mode, and these habitual patterns are a big part of the behavioral addictions that can take place in the most extreme situations. The habitual aspect of any addiction, like smoking or alcohol, makes quitting especially difficult, as multiple harmless triggers become associated with addictive action. Similarly, this is why the more we scroll through short videos the harder it becomes to avoid the same behavior the next day.


The truth is, dopamine is not really our enemy, and frankly, we have dopamine to thank for the survival of our species, but unfortunately tech companies use the underlying hackable mechanisms to advance their interests and create addictive patterns. However, the same system that has been used for centuries by our ancestors to survive and thrive can still be used today by mindful practitioners.

A certain amount of dopamine is always circulating in our brains, and it is the baseline that determines how we feel generally, but when the spikes in dopamine occur frequently, in response, baseline levels increase, as a result, the difference between baseline and peaks decreases. The level of difference to the baseline is what makes things feel exceptionally rewarding and lowering this range leads to not being excited and satisfied by things that otherwise were motivating and meaningful to pursue. This directly affects how we feel, our perception of energy levels, confidence, motivation, and desire to do anything.

Furthermore, dopamine is directly correlated with motivation to act on a desire. Multiple studies on rats as published in the journal of Neuropsychopharmacology (2016), also published in The Nature, show that when one rat is artificially depleted of dopamine and food is given to it, the rat still loves the food, but when the same food is placed just a couple centimeters away from the rat, it lacks the motivation to move and eat it. Additionally, if you ever had a feeling like you are stuck to the couch overeating snacks this might alert you, but a similar result was observed in rats when dopamine was artificially lowered, although the motivation to "work" for food significantly decreased, their free food consumption had increased.

For this and a multitude of other reasons I believe we should respect our individual dopaminergic systems, and treat our dopamine levels with caution. This reward and behavioral mechanism has been evolving before we formed as hominids, a similar mechanism is observed in other animals and it functions almost exactly like in humans. It is a system that is necessary for our survival, but unfortunately, it has not been encoded for the 21st century and it isn't perfect. Just like the limbic part of the brain, it is vulnerable to being hijacked and hacked by the modern environment.


To reiterate, although dopamine is not bad or evil, relying on maximizing dopamine response as the primary mechanism for retaining users on a platform can be harmful and negligent. The problem with this method as with every substance that produces a high amount of dopamine is that consistent dopamine peaks lead to increased benchmark levels of dopamine, so other actions would now have to compete with the benchmark for us to have the motivation to do them. The problem with substances that produce high levels of dopamine is exactly the same, after the addiction is in place no amount of real-world stimuli can generate as much dopamine as the drug, therefore an addicted person has no motivation to do other things and finds solace in drugs. Although not as extreme as substances, and not quite as damaging, the peaks caused by high volumes of short video consumption on social media over time can compete with the results of substance abuse in aggregate. This is why it is no surprise that more hours spent on social media platforms is statistically significantly related to higher rates of depression, although there are intricacies, the underlying mechanism and result could not be clearer.

Unfortunately, actions to limit this harm are negatively correlated with corporate goals of organizations that try to maximize time spent, for this reason, it is viewed as the necessary byproduct or the Cost of Doing Business. When it comes to adults this mechanism could be justifiable, after all, every pleasurable action we are involved in involves dopamine, and pushing stoicism on everyone is not constructive. However, in the case of teenagers who do not even necessarily have high purchase power, and are not the main targets of most advertisers, there should be more steps to limit the negative impact. A key takeaway would be that, although deleting social media platforms sounds tempting, we already rely heavily on them for work or personal use cases, and deleting them would not be realistic, or the best option, for most people. The solution that seems feasible is becoming more mindful of our and children's content consumption in an attempt to find a useful balance and prevent us from being stuck in behavioral loops.

This was an overview of the short video format; below are some other interesting facts that you might be interested in to let our minds wander and explore interesting subjects :)


Dopamine levels are key for learning, in our brains, an associative learning system that interprets stimuli roughly in "bad" and "good" categories has been set up, and when novel input is detected in the brain dopamine system fires up. A 2020 (Morrnes) study suggests that this dopamine response is crucial for learning and memory and if the dopamine response is blocked (tested on mice) the learning of the new stimuli is significantly slowed. This information is crucial as it suggests that the utilization of dopamine is crucial for learning and memory for humans. Accordingly, gamification (or other reward mechanisms that produce a dopamine response) of learning might be the best approach to learning, and we will be seeing more of it in the future.

Interestingly, when the studies of sports performance and coaching methods have been analyzed, they once again reinstated a well-known idea, that praising an athlete when he does something (even a small thing) right has much better results for performance and increased learning, than actively criticizing them. A certain level of criticism is necessary, but criticism-oriented teaching or leading approach can actually have a detrimental impact on a trainee, as criticism triggers a fight or flight response that stalls learning (HBR, 2019). It is true that evolutionarily strong negative emotions have been wired to make sure we remember not to do certain things or go to certain places that we experienced to be dangerous, but the same emotions and brain response would not contribute to learning what to do. So roughly the mechanism translates into - remember to do and go places that feel good and stay away from things that don't.

If you like this article, you might be interested in "Influencer Marketing of Rennaisance" Don't forget to check out the collection on Opensea: For everyone interested in neuroscience and how our brains work and interact with modern environments, I would strongly recommend Dr. Andrew Huberman's YouTube Channel


Costa, V. D., Tran, V. L., Turchi, J., & Averbeck, B. B. (2014). Dopamine modulates novelty-seeking behavior during decision-making. Behavioral neuroscience, 128(5), 556–566. Frank, S., Veit, R., Sauer, H. et al. Dopamine Depletion Reduces Food-Related Reward Activity Independent of BMI. Neuropsychopharmacol41, 1551–1559 (2016).

Cue-Evoked Dopamine Promotes Conditioned Responding During Learning.

Joachim Morrens, Çağatay Aydin, Aliza Janse van Rensburg, José Esquivelzeta Rabell, Sebastian Haesler, Neuron. 2020 Apr 8;106(1):142-153.e7. doi: 10.1016/j.neuron.2020.01.012. Buckingham, M. Goodall, A. (2019). The Feedback Fallacy. HBR

Conghui Su, Hui Zhou, et al. (2021) Viewing personalized video clips recommended by TikTok.

NeuroImage, Volume 237, 118136, ISSN 1053-8119,


Kalivas, P. (1993). Neurotransmitter regulation of dopamine neurons in the ventral tegmental area Brain Research Reviews, 18 (1), 75-113 DOI: 10.1016/0165-0173(93)90008-N

Kazuki, E, Naoyuki M, Sadamu N, et al (2011). Proceedings of the National Academy of Sciences. Note: The citation format is not quite academic as it is intended for the blog and not for publication. The information presented in the article is to the best of my current understanding of the media systems and is not for a medical disgnosis/advice

191 views0 comments

Recent Posts

See All


bottom of page