You On AI Field Guide · Phantom Flashbulbs The You On AI Field Guide Home
TxtLowMedHigh
CONCEPT

Phantom Flashbulbs

Neisser and Harsch’s discovery that vivid, confident memories of shocking events can be completely false—the human prototype of AI hallucination, and proof that in any constructive system, confidence and accuracy are independent.
Phantom flashbulbs are Ulric Neisser’s name for a specific and devastating failure mode of human memory: the confident, vivid recollection of an event that is, on verifiable evidence, simply wrong. The term emerged from a landmark study conducted with Nicole Harsch: the morning after the Challenger space shuttle exploded in January 1986, Neisser had students write down exactly how they heard the news, where they were, what they were doing. Nearly three years later, the same students were asked to recall the same event. Many accounts diverged wildly from the contemporaneous written record—different locations, different companions, different sources—and the students were supremely confident in these later, demonstrably false memories. Some, shown their own handwriting, insisted the original record must be wrong. The vivid, photographic feel of the memory—the quality that makes flashbulb memories feel like recordings—was no guarantee of accuracy; confidence and truth had come apart. Neisser’s conclusion was that memory is constructive: it does not replay stored recordings but rebuilds the past from
← Home0%
CONCEPTBook →

Keep reading with YOU ON AI

Unlock the full book, 10,000+ field-guide entries, and a 1000+ thinker library. If you have a book code, register now — it takes a minute.

Register with book code Sign in