Attention Splitter
A piece on attention economics
For those unfamiliar with the concept of the “IQ Shredder”, follow the hyperlink. It’s a necessary prerequisite.
The IQ shredder problem threatens human technological civilization, (or at least specific states), but it does so in a manner that seems tantalizingly simple. If one can fix the fertility problem, the IQ shredder is not a significant threat. This, of course, is easier said than done, but it is almost certainly possible. At the very least people are proposing solutions.
The Attention Splitter
Attention splitting is different. Attention is a vague idea, it isn’t exactly measurable. Technocapital doesn’t like ephemeral concepts; it likes numbers. There is a great deal of incentive to prove that a particular item of information is receiving attention stemming from monetization. Advertiser interest is directly correlated with the amount of attention an item receives. Because of this, ‘information sources’, (anything which attention can be paid to), began searching for ways to quantify the attention that they received.
At first, attention was measured in numbers of attendees, readers, etc. Then, advertisers discovered that some ‘attendees’ were not viewing the full content, meaning that if their advertisement was at the end of said content, it was not being seen. This incentivized ‘information sources’ to measure attention in units of time rather than the number of ‘attendees’. If you could demonstrate that your viewers spent more time on your content than your competitors, you could receive more advertisers. This method is still in use today, but it has a strong competitor: ‘clicks’.
Clicks, likes, views, etc. have an advantage in the short-form/image based media sphere of today. With the majority of ads front-loaded on content, it matters less if an ‘attendee’ views the full content than if they interact with it at all. These systems also hijack humans’ social functions—humans want to ‘like’ their friends’ content, and they tend to view internet celebrities as friends. This incentivizes people to spend their attention using emotion as well as desire.
The reduction of attention measurements to the smallest possible form has an unfortunate side-effect: it also splinters attention itself. As information sources compete more and more for small bits of human attention, humans find themselves less focused on individual pieces of information than on consuming small bits. Simultaneously, the amount of information available has increased exponentially. Humans’ capacity to consume information, however, has not increased. In other words, more and more people are spending their limited attention on small bits of information, much of which is useless.
This system has had a variety of effects on online content, (all negative). The one this essay is concerned with is the effort to grab attention from all sectors. This preexisting human drive is exacerbated by content monetization to become something of a factory farm for short-form content, even more so now that A.I. can create it. The influx of short-form content has reduced the time spent on any particular item of focus shorter and shorter on average, adapting individuals to an attention cycle which makes the consumption of longer or more complex content much more difficult. Human beings only have a finite amount of attention that they can ‘pay’ at any given point, and this attention is constantly being sapped by the digital platforms at one’s disposal. Participation is no longer voluntary—TikTok is on gas station pumps now.
The Threat Posed
“In an information-rich world, the wealth of information means a dearth of something else: a scarcity of whatever it is that information consumes. What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.” -Herbert Simon, (pulled straight from Wikipedia).
This is a reversal of how we normally treat information. Rather than information scarcity, Simon correctly points out that what we lack is the attention to ‘pay’ for all of the information available.
If we treat information as a consumer of attention, (as explained above), we quickly conclude that A), a limited amount of information can be consumed in the first place and B), the fragmentation of attention into smaller segments reduces the amount of important information which will be consumed. This makes the reduction of attention a serious problem.
Human technological civilization requires a certain level of important information to be consumed in order to sustain itself. While the majority of humans spend their excess attention on leisure, this wasn’t an issue. Intelligent humans tend to spend their leisure time consuming information anyway. In modernity, however, we require a much higher portion of the population to consume information to function properly.
Of course, progression has reduced humanity’s dependence on “primary knowledge” before. For example, mechanics aren’t building their tools, and the factory isn’t designing them. With the introduction of writing, information could be stored and accessed only when needed, doubly so with the internet. This is an example of an augmentation technology: something that reduces the amount of information an individual must know. A simple example is Youtube tutorials on vehicle maintenance. As augmentation technology improves, the amount of information which must be remembered ‘manually’ will decrease to some extent. We will return to this when I discuss A.I.
Putting A.I. and augmentation technologies to the side for now, we realize that the current level of human industrial civilization may actually be the limit of what can be achieved. Where we are now, the technologies available are continuously hijacking your attention; this is the express purpose of social media applications for example. It is doubtful that these applications are going anywhere, and even if they did, something functionally equivalent would likely replace them.
Just as the IQ shredder has reduced human technological capacity at various points in our history, the attention fragmenter may present a ceiling of sorts for human development. As long as technocapital incentivizes the translation of attention into the kinds of currencies that it does, it is unlikely that the overall amount of attention will increase significantly. Population may increase, but the total attention is unlikely to with more people interacting with modernity. IQ is also correlated with the value of attention: the more intelligent one is, the more valuable the information which they can process.
The attention splitter is happening to all humans living in industrial or “post-industrial”, (to the extent that it exists), society. Every individual is consistently being bombarded with an excess of information, the majority of which is useless. The humans living in these societies are the most intelligent on the planet, so their attention is worth more, yet it is being spent on less valuable information most of the time. Human society at this level, however, requires the consumption of more information than ever before to sustain itself. Functionally this means that human society as it stands requires at minimum the current level of attention. Realistically, it probably needs the level of attention that was available in 2005 at minimum.
This issue is exacerbated by two facts: the limits of intelligence and the continuing expansion of technological civilization. Technological civilization is expanding into what we can politely call the “developing world”. The regions it grows the fastest in have the least intelligence to spread. In this geographic respect, the amount of information which needs to be consumed is rapidly outpacing the amount of attention available.
This is happening at an even grander level within the most developed states: both intelligence and attention are decreasing while the civilization is still becoming more complex, (albeit slowly). What this means is that at some point, instead of gaining information, civilization will be collectively losing information. This information will likely be preserved in some literary format, but if information remains unconsumed it functionally does not exist.
Outcomes
There are three possible outcomes to this scenario as I have laid it out.
Humanity stagnates at a relatively similar level of technological civilization to where we are now. There is not much further that attention levels can really be divided; the current manners of their measurements are already ‘instant’. However, this level has already demonstrated its consequences. Humans exposed to modern technocapital systems, (which are still growing geographically), will continue to have shortened attention spans. As a result, they will consume less and less information-dense content. This lack of the consumption of important information for the maintenance of the underlying systems means that these systems are limited in how far they can expand. Technical systems, educational facilities and research capability will all be greatly limited. Of course, this is not a recipe for decline either—the first thing to decline in these systems would be their mass-media apparatus, but it is a limit for progression.
Artificial intelligence does not have the same limitation. If A.I. in its ‘true form’, (A.G.I.) is eventually created, it will obviously be capable of consuming a greater degree of information. This guarantees the replacement of humanity in a great number of fields. This would enable significantly more progression than outcome one, but with the caveat that it relies on artificial intelligence. At the worst, this outcome results in singularity.
Humanity divides between a small number of individuals who ‘unplug’ to a certain degree. This new elite is able to maintain their attention and intelligence despite the influence of technocapital. These individuals would have a disproportionate amount of power in such a world.
None of these outcomes are mutually exclusive. Obviously the majority of readers will view the third of these as a breath of fresh air in what has been quite the pessimistic essay. However, optimism may be a bit premature considering the reality of the situation.
The Majority
What we have seen in the past two centuries is that civilizational progress is a secondary goal for the majority of humans. Welfare in one form or another is much more popular. As the attention of the majority declines and cognition goes with it, dependency on the ever increasing welfare state will increase. Post-Marxist ideas of wealth combined with identity politics, (incensed by the natural characteristics of cognition becoming more apparent than ever), will remain a popular ideological mix.
As a result, the ability of this ‘elite’ discussed in outcome three to generate technological or civilizational progress will be hampered. Calls for wealth redistribution will increase with dependency and population increase. Short-form content enables every human with a smartphone, (soon to be all of them), to create propaganda from their bedroom, reducing the ability of the ‘high'-functioning’ elite to counter populist messaging.
Natural Trajectory
Once again placing A.G.I. to the side, the outcome is pretty obvious. Technological civilization will remain similar to where it stands with a small subset of the population focusing on maintaining their cognitive abilities and focus. The rest of the population will embrace left-wing populism more and more each day, squeezing the small subset which generates wealth further and further. Eventually this system will become unsustainable, leading to welfare-induced collapse.
High-functioning A.I. would delay this outcome. However, only A.G.I. could switch the trajectory away from slow decline. This, however, risks producing other issues.
Lastly, this small elite may realize that there is only one way that they can prevent collapse and avoid a total A.G.I. replacement. This is the bionic horizon.