Collective Attention

In the AI Story

The classic examples of collective attention are moments of crisis or celebration that commanded near-universal focus: the Moon landing (600 million simultaneous viewers), the fall of the Berlin Wall, the September 11 attacks, the early weeks of the COVID-19 pandemic. In each case, the convergence of attention was enabled by media scarcity—limited channels, shared broadcasts, common information sources. The scarcity was a constraint (gatekeeping, limited perspectives, manipulation) and a condition for collective attention (everyone watching the same three networks meant everyone encountered the same information). The digital transformation replaced scarcity with abundance—infinite channels, infinite sources, infinite personalization. The constraint was lifted. The condition for collective attention disappeared with it.

Citton's ecological framework reveals that collective attention is not merely a side effect of limited media options but a positive achievement requiring specific institutional support. The nightly news broadcast was not just information delivery—it was a collective attention ritual, a daily synchronization of focus that created the shared informational ground on which next-day public conversation depended. When that ritual disappears—when personalized feeds replace broadcast schedules, when AI-generated summaries replace shared articles—collective attention loses its temporal and informational anchors. The society can still respond to crises, but the response is fragmented: each faction attends to different evidence, interprets through different frames, and coordinates only within algorithmically similar subgroups. The whole can no longer focus as a whole.

AI accelerates fragmentation through two mechanisms. First, it generates the flood—producing content at volumes that exceed any collective capacity for shared processing. When a million new articles, images, and videos appear daily, the society cannot jointly attend to any significant fraction of them—the cognitive bandwidth for collective attention is finite, and the supply has become functionally infinite. Second, AI personalizes within the flood—ensuring that each individual's path through the deluge is uniquely optimized, which guarantees that no two individuals traverse the same informational landscape. The convergence that collective attention requires becomes statistically improbable. The commons evaporates not through demolition but through infinite subdivision into personalized micro-niches.

The reconstruction of collective attention under AI-abundance conditions is the central governance challenge Citton's framework identifies. If scarcity created collective attention as an accidental byproduct, abundance requires it as a designed goal. This means building attentional institutions—platforms, practices, policies—that deliberately convene collective focus. Public media producing content designed for shared attention rather than individual capture. Digital town halls creating occasions for synchronized focus on common questions. Educational systems teaching the capacity for joint attention (discussion, deliberation, collaborative inquiry) as a core civic skill. The institutions will not emerge from market forces—personalization is more profitable than shared focus. They require political will, public investment, and the recognition that collective attention is a commons requiring active governance, not a market outcome to be left to algorithmic optimization.

Origin

Collective attention as a political concept has roots in Émile Durkheim's collective effervescence—the heightened emotional and intellectual state during shared rituals—and in Walter Lippmann's analysis of public opinion formation through shared media. Citton's ecological extension treats collective attention not as a ritual state but as an ongoing practice requiring continuous environmental support. The innovation is recognizing collective attention as a commons—a shared resource that can be depleted, that requires governance, and whose degradation produces specific, diagnosable social pathologies (polarization, institutional distrust, inability to coordinate responses to collective threats).

Key Ideas

Emergent, not additive. Collective attention is not the sum of individual attention but a qualitatively distinct phenomenon—a society thinking together, requiring convergence on common objects.

Media-dependent infrastructure. Collective attention requires specific media configurations (shared broadcasts, common texts, synchronized events) that algorithmic personalization systematically eliminates.

Democracy's substrate. Democratic deliberation presupposes collective attention—citizens must attend to the same problems and evidence to deliberate meaningfully about them.

Fragmentation as default. In AI-saturated environments, collective attention does not emerge naturally but must be designed—deliberately constructed against the optimization logic that fragments focus.