There is a well established principle in nutrition science that the brain takes roughly 12 to 20 minutes to register that the stomach is full. The phenomenon even has a name: sensory specific satiety, first described by French physiologist Jacques Le Magnen in 1956 and formally defined in 1981 by researcher Barbara J. Rolls. The principle is simple but profound, the pleasantness of a stimulus already consumed drops significantly, while the appetite for something new remains unchanged. Every dietician draws the same conclusion: eat slow, chew more, pause between bites. The signal is coming. The body needs time to receive it.
I thought about this recently not at a dinner table, but while scrolling through my LinkedIn feed. I had opened the app with clear intent, to catch up on industry news, see what peers were sharing, stay current. Within three minutes I was seeing the same five people. Again. Out of a network exceeding a thousand connections, the feed had quietly narrowed to a loop. I kept scrolling not because I was finding value, but because the “I’ve had enough” signal had not arrived yet.
It did not stop there. On a news aggregator platform I follow, I encountered the same article, not a similar story, the identical piece, surfacing multiple times across a single month. On Facebook Reels, actively swiping in search of something fresh, the same reel reappeared within the same week. The algorithm, unable to distinguish between content I had noted and content I still needed, kept serving what it had already served.
That pattern of recognition is what this article is about. But the implications go considerably further than one person’s scroll behaviour.
No platform currently measures cognitive presence. They measure time. They measure clicks. They measure completion rates. Not one reports whether the human behind the metric was actually there.
The Science Behind the Scroll
The human brain does not evaluate content in real time. There is a biological lag between consumption and the recognition of consumption. Neuroscientific research confirms that advertisement fatigue is rooted in the habituated response of neurons in the prefrontal cortex, amygdala, and hippocampus, the regions responsible for attention, emotion, and memory. Once these regions register sameness, engagement declines rapidly and no amount of additional exposure reverses that drop.
Platforms have built their entire engagement architecture on exploiting the delay before that signal arrives. The algorithm serves content. The user consumes it. Before the brain can assess whether that consumption was satisfying, the next piece arrives. The scroll is engineered to outpace the brain’s evaluation cycle, locking users in a permanent state of almost full but not quite.
Peer reviewed consumer research published in 2025 goes further, identifying what scientists now call cognitive rigidity, a state in which homogenised algorithmic content over occupies attention and cognitive energy to the point where the brain loses its capacity to receive anything different. The algorithm does not merely tire the user. It progressively narrows the mind’s aperture. Sensory specific satiety captures the fatigue. Cognitive rigidity captures what follows.
Platform by Platform: The Same Failure, Different Surfaces
Facebook was the first platform where this failure became visible at scale. Despite users carrying networks of hundreds or thousands, feeds began to feel claustrophobically narrow. Mark Zuckerberg acknowledged in 2025 that time spent across Facebook and Instagram had declined meaningfully. The algorithm, designed to serve what previously engaged you, created a feedback loop so tight it became a trap. The logic meant to retain users began, quietly, to exhaust them.
LinkedIn presents the same structural failure with a professional veneer. Views on the platform are down 50%, engagement down 25%, follower growth slowed by nearly 60%. The platform’s own data showed users spending less time despite seeing more high engagement content. The algorithm rewards prior engagement, surfacing the same cluster of voices, in the same formats, session after session. A precision engine systematically narrowing the world it promised to open.
Short form video platforms compound this with speed. The algorithm queues similar content the moment you finish watching. For two to five minutes the dopamine loop holds. Then the novelty signal fades. The fifth consecutive reel on the same theme triggers a response the platform has no mechanism to detect: the user is still scrolling, but the brain has already left. YouTube’s autoplay function replicates this across longer content, watchtime metrics climb while cognitive engagement quietly exits.
The result, confirmed by a 2026 industry report, is what researchers are now calling “scroll immunity” audiences developing an automatic skip response, present in session data and absent in every meaningful sense.
When the Algorithm Overrides Your Identity
India presents a dimension of this failure that global commentary has entirely missed, and it goes beyond repetition into something more troubling: identity override.
A viewer who actively watches content across five languages, understanding each culturally, not merely linguistically, will find the feed defaulting mid-session to auto dubbed versions of a dominant language. Tamil reels giving way to dubbed Hindi. Regional language content displaced by an algorithmically assumed preference. Meta launched AI-powered auto dubbing for Facebook and Instagram Reels in August 2025, automatically translating and lip syncing content into other languages at scale, a platform feature deployed without asking users whether they want their linguistic identity overridden in the name of engagement optimisation.
The algorithm registered watch time and concluded that language preference is fungible. It is not. Language is not a content filter. It is cultural identity. When a platform overrides that identity mid session, it has not just served the wrong content, it has mistaken who you are. And the advertiser who bought a Tamil speaking audience found their message delivered in a language that audience did not choose.
News Aggregators: When Repetition Erodes More Than Attention
India’s news aggregator ecosystem adds yet another dimension. These platforms serve the same story repackaged across multiple cards, sources, and formats within a single session. The brain registers repetition even when the byline changes. Five cards on the same development, served in succession, produce the same satiety response as five consecutive reels on identical content.
The difference is that news fatigue carries an additional and more damaging cost: it accelerates cynicism. Users who repeatedly encounter recycled coverage with no new information do not conclude the platform is redundant. They conclude the news itself is. The aggregator algorithm, optimised for engagement, is quietly degrading the perceived value of the journalism it distributes. This is not a side effect. It is a structural consequence of optimising for repetition over relevance.
The Programmatic Precision Paradox
Here is where the advertising industry must confront a truth it has so far managed to avoid.
Programmatic advertising was built on a promise: reach the right person, at the right moment, with surgical accuracy. The industry has invested billions in data infrastructure, audience segmentation, bid optimisation, and attribution modelling. We can target by age, income, intent, device, language, and geography. We can bid on a specific impression in under 100 milliseconds. Dentsu’s DC Trends 2025 report calls algorithmic fragmentation a transformative force reshaping brand strategy. eMarketer confirms diminishing returns flattening growth across every major social platform.
And yet the industry has no mechanism to determine whether the person targeted was cognitively present or cognitively rigid, when the content arrived.
The perfectly targeted ad is routinely landing on a brain that has already processed fifteen similar stimuli in the preceding four minutes. Programmatic precision met cognitive absence. The impression was served. The human was not there. Neuroscience confirms this costs real money: participants who encountered relevant ads in an engaged state showed approximately 27% higher brand recall than those reached in a fatigued condition. The gap between a present audience and an absent one is not marginal. It is measurable, significant, and currently invisible to every dashboard in the industry.
This is not a technology limitation. The signals exist, scroll velocity, session depth, interaction latency, content variety indices, all could model cognitive fatigue in real time. The reason they are not used is simpler and more uncomfortable: measuring audience fatigue would devalue inventory. And devalued inventory is a conversation no platform is incentivised to initiate.
What the Ecosystem Is Losing
For content creators and publishers, the consequences are structural. Your most carefully considered piece enters a feed that has already exhausted its reader. The algorithm did not bury you because your work was weak. It buried you because it fed the reader too much of what looked like you before you arrived. The fatigue is pre-loaded. The opportunity is already spent.
For brands, the situation is equally stark. A brand that reaches the right person with the right message at the wrong cognitive moment has not achieved precision. It has achieved waste at scale, with a detailed attribution report to make the waste look productive. Research confirms that users aware of algorithmic manipulation exhibit inconsistent resistance, oscillating between active pushback and passive compliance. They are scrolling. They are not buying. They are not remembering. They are present in the data and absent from the outcome.
The algorithm will not self correct. No regulator has the framework to intervene. No advertiser consortium has the leverage to demand transparency on cognitive presence. The system will continue to optimise for the signals it is given, and the ecosystem will continue to purchase impressions it cannot verify.
In a world of programmatic precision, the most valuable data point we do not collect is the simplest one: is anyone actually home?
The algorithm forgot you were full. The industry forgot to check.
About the Author
GV Krishnamurthy (GVK) is Partner at AdNexa.ai and a veteran media strategist with over three decades of experience spanning broadcast, print, digital, and advertising strategy across India. A committee member of the Advertising Club Bangalore, GVK writes regularly on media accountability, platform behaviour, and structural shifts in India’s advertising ecosystem. Views expressed are personal.
(Views are personal)
