Retractions fail partly because the brain engages model-updating less when corrections challenge prior beliefs

Evidence

The authors found that the retractions elicited less activity in the right precuneus, an area of the occipital lobe. Because this brain region has previously been implicated in memory processes and difficulties in integrating information, the authors concluded that basic integration and coherence mechanisms are less active when new information challenges people’s previous understanding of an event that requires encoding, consistent with the ‘model-updating’ explanation. Of course, it’s good to keep in mind that the study only included twenty-six volunteers and initial support for one account doesn’t rule out other plausible theories. In fact, a follow-up study from the same research team suggests that the problem also likely stems from the challenge of having to hold both correct and incorrect information in memory concurrently (which is more aligned with the ‘selective retrieval model’). — Sander van der Linden, Foolproof, 279

Categorization

• QUESTION: The two competing neural models (model-updating vs. selective retrieval) both remain live; a definitive test separating them would be worth chasing.
• XREF: connects to: Misinformation leaves residual traces in memory even after correction — Sander van der Linden, Foolproof: Why Misinformation Infects…, 241; Awareness of a correction does not stop misinformation from infecting related memories; Motivated reasoning engages different brain circuitry than neutral reasoning. — Lee McIntyre, Post-Truth, 103 (+2 more)

Broader Connections

• Awareness of a correction does not stop misinformation from infecting related memories #corroborates (0.85)
• Hidden biases can be corrected by training attention, not by adjusting conditions #responds-to (0.75) — Jenny Odell, How to Do Nothing, 394
• Stress drives the brain toward cognitively cheaper confirmation bias #corroborates (0.75) — Sander van der Linden, Foolproof, 117
• Motivated reasoning engages different brain circuitry than neutral reasoning. #corroborates (0.72) — Lee McIntyre, Post-Truth, 103