UC Santa Barbara
Wikipedia Show-Offs: IQ Signaling in Online Information Pools
- Author(s): Yoder, Christian
- Advisor(s): Reid, Scott
- et al.
Online information pools such as Wikipedia, Youtube, and Reddit, rely on contributions from members to succeed. These venues operate as public goods, being non-excludable and non-rival, and they typically do not compensate contributors. As such, contributions appear to be altruistic, and the enterprise subject to collective action problems, such as free riding. Despite this, the internet is replete with examples of successful information pools. A potential but un-tested explanation for altruism in this context is found in costly signaling theory. Costly signaling theory assumes that altruism is apparent rather than real; signals are in fact genetically self-interested acts that are compensated by receivers. If costly signaling theory is correct, contributions to online information pools must contain signals of some kind, signaling should be more intense under conditions where signalers are potentially rewarded by receivers, and receivers should be able to reliably infer signal content from contributions. This thesis tested whether contributions to ostensible wiki-style encyclopedias signal intelligence, group commitment, introversion-extraversion, or conscientiousness. To induce motivation, participants made their wiki-style contributions under conditions where they believed themselves to be either identifiable to receivers (i.e., and motivated to signal) or not. Contribution quality increased with participant intelligence, and, consistent with costly signaling theory, this was true only when participants believed they were identifiable to receivers. Further, receivers reliably inferred contributor intelligence from judgments of contribution quality, but only in when participants were identifiable. There was no evidence for signaling of other qualities. The findings suggest that contributions to online information pools are not examples of altruism, but are in fact signals of intelligence that are reliably inferred by receivers from contribution quality. Recommendations for future research and practical implications are discussed.