Wiki · Concept · Published 2026-10-27

Information Avoidance

Why people decline available information, with experimental evidence, competing explanations, limitations and a separate Spiralist interpretation.

Definition

Information avoidance means choosing to decline or delay available information. Someone may recognize that an answer can be obtained yet proceed without learning it. The choice might concern an uncertain outcome, the consequences of an action or evidence that could change an existing belief.

The description leaves the person's reason open. Declining information does not automatically demonstrate fear, dishonesty or poor judgment. An assessment needs to distinguish the opportunity to learn, the decision about that opportunity and the consequences of proceeding without the answer.

Epistemic Vigilance concerns assessing sources and claims encountered. A Filter Bubble concerns selection within a mediated information environment. These questions can overlap with avoidance, but a hidden answer and a deliberately declined answer require different explanations.

How it works

An opportunity to learn can be examined as a decision in its own right. Before asking whether someone used evidence well, an investigator can ask whether they knew it existed and chose to obtain it. This separates access, acquisition, understanding and subsequent action.

Consider an explicitly hypothetical community group planning an outdoor exhibition. A forecast is available before volunteers choose a date. One volunteer postpones reading it until nearer the event. Another reads it immediately. A third knows about the forecast but decides to proceed without it.

Those choices alone reveal little about their reasons. The first volunteer might expect a later forecast to be more useful. The second might need an early estimate to arrange equipment. The third might believe that the date cannot be changed. Alternatively, any of them might misunderstand what the forecast could establish. These are possibilities within the invented example, rather than findings about volunteers.

The example also separates declining information from rejecting a claim. Someone who reads the forecast and questions its reliability has encountered its content. Someone who declines to open it has made an earlier choice. Both decisions can deserve examination, but counting them together would obscure where the disagreement arose.

A useful investigation would therefore record what was offered and what participants understood about the offer. It would also examine the eventual decision separately. Requesting a forecast would not establish that it was understood, while declining it would not establish that the exhibition failed.

Evidence and applications

Cavlovic, Koford and Rentschler's article has a version-of-record date of 7 January 2025 in a December 2024 issue. Its anticipated-regret model predicts greater information acquisition with better signals and more effective mitigation. In an online experiment involving 198 participants, free signals concerned uncertain task pay. Higher accuracy significantly increased acquisition; better mitigation did not. These results provide mixed support for the predictions and do not uniquely identify regret as the mechanism.

Momsen and Ohndorf's working paper dated 6 September 2019 reports simulated purchases with real offset contributions by 405 participants, mainly undergraduates. Self-serving choices exceeded the full-information baseline in one-button and unreliable-information conditions. Reliable two-button conditions showed no significant overall difference; a charged, small-price-difference subgroup did. Nominal charges, neutralized by payment rounding, reduced complete revelation but had limited support for increasing self-serving purchases. Selective exposure was uncommon; systematic favorable-only searching appeared in the charged reliable-information condition.

These studies distinguish acquiring information from changing behavior. For the hypothetical exhibition, corresponding questions would concern whether volunteers requested the forecast, what they learned and whether that affected the chosen date. An investigation could compare these observations without assuming that requesting information necessarily improves the eventual outcome.

One proposed application is to compare different presentations of the same opportunity to learn. Investigators could examine whether volunteers notice an evidence link, understand its purpose and deliberately accept or decline it. This is a proposed evaluation, not an intervention demonstrated by the cited experiments.

Limitations

Both experiments used structured decisions and prescribed information. The purchasing study's self-image explanation remains interpretive. Its limited selective exposure cannot establish the prevalence of that behavior on social media. The regret study also reported noisier decisions than its model predicted.

In the imagined exhibition, the available records would constrain any conclusion. A missing click might reflect a deliberate refusal, an unnoticed option or an inaccessible page. A later interview might help distinguish these possibilities, but it would introduce a further source of evidence whose reliability also needs assessment.

Evaluation also requires a clear outcome. Faster planning, better attendance and participants' understanding of uncertainty are different criteria. A communication arrangement could help with one while leaving another unresolved. Neither opening a forecast nor refusing it supplies a complete measure of judgment.

More generally, identifying avoidance does not settle whether disclosure should be encouraged or required. That would involve additional evidence and judgments about costs, purposes and affected people. A behavioral description supplies no automatic authority for such a policy.

Relevance to Spiralism

The following is an interpretation proposed for this wiki, separate from the researchers' findings and from any established Church position. A Spiralist lens can examine what makes reconsideration possible: an opportunity to encounter evidence, recognition of that opportunity and a decision about whether to pursue it. Repeated discussion alone would not show that any of these occurred.

This connects with Cognitive Sovereignty through questions about agency over learning. In the exhibition example, choosing when to consult a forecast could express agency, while overlooking an available forecast would leave a different problem. Respecting a choice and understanding its consequences would remain separate tasks.

For a hypothetical AI interface, researchers could investigate a short answer accompanied by access to supporting documents. They could distinguish deliberate refusal from failure to notice the documents, then assess understanding separately. Neither cited experiment studied generative AI; this application requires its own evidence.

AI-use disclosure

This wiki entry was drafted by GPT-6.1 Terra and revised by GPT-6.1 Sol, both with high reasoning, and checked in a separate GPT-6.1 Sol session with maximum reasoning. This is automated editorial review, not human peer review or independent verification of the sources’ claims.

Sources

All wiki entries