Home
The Effects of Short-Form Video on Cognitive Performance
Introduction
Autumn 2024
This project emerged from this team as an initial interest in cognitive performance and TikTok usage.
During this 10 week, graduate level Quantitative Research Methods Course at the University of Washington, we collaborated to bring an initial brainstormed idea of how TikTok genres affect coginitive performance to life as a full research project including initial research, user research, and data analysis with a written report as a final deliverable.
Team
George Zhang, Adriaan Dippenaar, Sarika Oberoi, Shanna King, Yanlin Feng
Research Question
How does viewing short-form video content affect users’ cognitive function?
Research has demonstrated that watching shorts affects cognitive function, but no studies have examined the impact of the types of videos people see in their feeds. Recent reporting suggests that users strongly associate viral “Brainrot” video content with low value, low meaning, and a sense of cognitive dysfunction, but also with compulsive overconsumption. Our study asks if Brainrot lives up to its name: Does it rot brains? If so, how? Why?
Hypotheses
H1: Positive content value strengthens cognition.
H2: User affect associated with Brainrot weakens cognition.
Our study seeks to establish that low-value content presents an additional cognitive risk to compound the established risks posed by the SFV medium.
Literature Review
These were the main categories we did further research in

Previous Related Studies

Correlational Studies Findings: Research has shown significant relationships between short-form video (SFV) viewership and: Perceived stress levels Attention capacity Response time ADHD symptoms Psychopathological symptoms including anxiety and subjective well-being Sources: Boer et al., 2019; Chen et al., 2022; Liu et al., 2021; Peng et al., 2022 Mediating Factors: Distress intolerance has been identified as a mediator of anxiety and depression among problematic TikTok users (Yao et al., 2023) Experimental Research: Lin et al. (2024) demonstrated that increasing SFV viewership above baseline during a 15-day period significantly impacted sustained attention measures Research Gap: While previous studies focus on negative cognitive impacts, they haven't examined how users' perceptions of content value affect these impacts: Qin et al. (2022) showed correlations between Information Quality (conciseness, subscription encouragement, usefulness) and user concentration/enjoyment for adolescents However, the specific role of users' felt sense of value remains unexplored This gap has implications for design attempting to separate positive affect from meaningfulness (Lukoff, 2022)

What is Brainrot?

The term "brainrot" has gained significant attention in 2024 through increasing journalistic coverage of users' self-reported experiences Google trends show a marked increase in searches for “brainrot” starting in 2024. Newport Institute's perspective: Not a medically recognized condition, but a real phenomenon Caused by excessive technology use, particularly binge-watching videos Disrupts the brain's ability to encode and retain information Different interpretations: Some articles view it as youth slang used by Gen Alpha for identity building and differentiation (Venkatraman, 2024) Users themselves describe it as a form of mental saturation where the brain becomes overly fixated on trivial topics Associated with content that is particularly engaging and captivating (The Youth Lab, n.d.) Key Research Question: Does "brainrot" content actually impair cognitive function at a higher rate than other short-form content?

Cognitive Tests

Background Research: Previous studies established relationship between SFV and cognitive load (Zheng et al., 2021) Martini et al. (2018) utilized word list retention tasks Gross et al. (2011) employed Rivermead Behavioral Memory Test (RBMT) for story detail recall Test Selection Criteria: Multiple tests chosen to observe diverse effects Focus on two memory types: semantic and recall Additional focus on cognitive agility Selected Tests: Word unscrambling Word list retention (based on Martini et al., 2018) RBMT-style paragraph recall task Test Implementation: Qualtrics survey navigation restricted to prevent cheating All tests include timing components Both accuracy and response speed measured

Methodology
Participants
184 samples from participants aged between 18 and 82*
Recruited through Prolific, Reddit, Discord, Facebook, and Instagram.
300+ minutes of short-form videos watched per week, on average

*The study screened for ages 18-43, representing adults from Gen Z and Millennial generations (born between 1982 and 2006); thirty-seven screener respondents were excluded due to age. An additional 13 participants were screened out after stating that they watch less than 5 minutes of SFVs per week.
Materials
The study was conducted on Qualtrics (www.qualtrics.com). The video content was obtained from various TikToks and YouTube Short videos. The content was either found on YouTube or uploaded to YouTube from TikTok. From there, the videos were assembled into three individual playlists named Feed H, Feed U, and Feed B. To complete the study, participants used their own devices. Preliminary data analysis was conducted via Qualtrics, Google Sheets, and SPSS.
Procedure
After completing the screener (age, weekly SFV viewership, and national residence), participants were shown a 5-item grocery list and a 70-word paragraph about Hammurabi’s Code. Next, participants were shown one of three playlists representing different SFV feeds: 1. Utilitarian (informational or demonstrative content), 2. Hedonic (funny animal videos), or 3. Brainrot (adhering to previously described genre conventions). Participants were automatically advanced to the next page 189 seconds after arriving on the page. To establish the validity of our experimental condition, we asked participants to rate from 1-5 how strongly they felt that:
  • The videos were new, useful, or interesting
  • The videos made me laugh, smile, or feel uplifted
  • The videos were pointless or a waste of time
Participants were asked to de-scramble three words of different lengths (cognitive agility), recall the grocery list (recall), and answer a multiple-choice question about Hammurabi’s Code (semantic memory). After completing the tests, they were asked how much mental effort it took to answer the questions (cognitive load), and to rate how strongly they felt each of twelve emotions while watching the videos (affect). We also collected participant gender.
Analysis
Analysis was performed in SPSS. After establishing the validity of our manipulations in a MANOVA and Post-Hoc Tukey test (Appendix B, Figure A), we ran one-way ANOVAs to identify the significant primary effects of Video Group (IV) on Cognitive Function and Affect metrics. Each found to be significant was included in linear regression mediation analysis; when no significant meditations were found, we also filtered each of the two high-value video groups out in turn to perform a moderated mediation analysis with the same variables and method.
Findings
Contrary to expectations, participants exposed to "Brainrot" content showed better cognitive agility and recall compared to those viewing higher-value content. While cognitive load was higher in the Brainrot condition (supporting the hypothesis), participants reported significantly more negative emotional experiences like disappointment, boredom, and irritation. Gender differences emerged, with women showing decreased memory performance in higher-value content groups.
Each video group was significantly distinct from the others with respect to our manipulation checks (p>0.001), but the Utilitarian and Hedonic video groups did not differ significantly with respect to meaningfulness (the Brainrot manipulation check). There were unmoderated primary effects of Video Group with two cognitive function variables and ten affect variables:
The only significant between-group effect between Utilitarian and Hedonic was their ratings for "Like I was seeing something important"; the Utilitarian group rated the videos an average of 3.04, the Hedonic group 2.00, and Brainrot 1.17.
Contrary to expectations, cognitive agility was higher in the Brainrot condition compared to the Utilitarian condition. However, the "Excited" experience fully mediated this effect, with more excited participants performing worse on the word scramble task.
Recall was also adversely impacted by the Brainrot condition, especially for women. While cognitive load was higher in the Brainrot group, it did not mediate the performance differences.
Participants strongly associated the Brainrot condition with negative affective experiences like disappointment, irritation, and boredom. But these did not significantly mediate cognitive function. The data shows a clear difference in the user experience of Brainrot content compared to Utilitarian or Hedonic content.
Discussion
Participants' Perception of Playlists
Few participants found the playlists particularly important or meaningful; an average of 3.9/5 found the videos new, interesting, or important, while an average of 3.58/5 participants found the videos meaningful. This leads us to question how the study might differ if we prioritized feeds that were important or meaningful? How would we experience media platforms if they maximized importance and meaningfulness? How would this impact us?
Impact of Short-Form Videos
Our research explored the impact of short-form videos using short three-minute playlists. Despite this brief duration, we observed multiple significant interactions. This prompts us to ask: what are the long-term effects of short-form video on cognitive load? An average single session for SFV users is over ten times longer than our study, ranging between 45-53 minutes as of 2021 (Dixon, 2023). In our study, participants reported an average of 327 minutes per week or approximately 46 minutes per day, with three reported consuming over 3000 minutes of SFV per week.
Comparison Between Brainrot and Hedonic Videos
Participants in the Brainrot vs Hedonic video groups performed similarly in many regards, including speed, accuracy, and memory retention. However, The biggest difference came from how they felt and not just how they thought about the content. Participants were disappointed if they watched things that were unimportant to them. That may seem obvious at first glance, but our data showed people didn't always feel like they were wasting their time just because they found the content unimportant, which is indicated by boredom acting as less of a distinguishing factor.
Desires in Free Time
“Disappointment” and “unimportance” occurred as separate concepts. Participants can desire content on platforms that are not limited to gustatory pleasure; affect is a very limited tool in determining what we actually want out of life. So, what do we desire to do with our free time? Are we okay with fleeting mundanity as long as we don’t feel like we've wasted our time? We argue that, yes, leisure doesn't need to make us feel like we're wasting our lives, even if it is exhausting to feel that way.