Semantic Decay Protocol: Memory and Corridor Traversal
Date: Summer 2019Subjects: 18Location: East Wing BDuration: 6 weeks
⚠ ANOMALOUS RESULTS
18 laminated word cards were placed at regular 2-metre intervals along the corridor. Subjects walked the corridor once and were immediately asked to recall as many words as they could.
Expected: standard forgetting curve — typically 7±2 items from a sequential list. Actual: subjects recalled between 12 and 31 words. The problem: there were only 18 cards.
Subjects were recalling words that were not placed in the corridor. The extra words were consistent across subjects — suggesting a shared source rather than independent confabulation.
Recalled Anomalous Words (appearing in 3+ subjects)
Word
Times Recalled
Actually Placed?
Subjects
CORRIDOR
14/18
NO
S01,03,04,07,08,09,11,12,13,14,15,16,17,18
RETURN
9/18
NO
S02,04,06,08,10,13,14,16,17
始 (shǐ — begin)
7/18
NO
S01,05,07,11,13,15,17
HOME
6/18
NO
S03,07,09,13,14,16
ALREADY
5/18
NO
S04,08,12,15,18
"They are not remembering the words I placed. They are reading the corridor itself. The corridor has a vocabulary. It has words it wants them to know. I am underlined twice: CORRIDOR, RETURN, ALREADY. These are the corridor's words. I don't know what it's trying to say — yet. — J."
CONCLUSION (REVISED):
Standard semantic decay models do not apply to this corridor. The corridor appears to be generating additional semantic content during traversal. Source: unknown. This experiment will not be replicated as designed — it cannot be: the corridor appears to learn from repeated exposure and the word list no longer serves as independent stimulus.
the corridor's message, fully assembled: CORRIDOR RETURN ALREADY HOME BEGIN — you were already here. you have already returned. this is the corridor.