Three expts

  1. infer goals given utterance and word-scale state
  2. infer goals given utterance and heart-scale state
  3. accept word (feeling) given heart-scale state

Expt 1 (most recent): goal attribution given utterance and heart-scale state

plot of chunk unnamed-chunk-4

Expt 2: goal attribution given utterance and word-scale state

plot of chunk unnamed-chunk-6

Expt 3: word (feeling) acceptance given heart-scale state

plot of chunk unnamed-chunk-8

Original vs. replication

Exclude expt3, which is about semantic distribution (should proximate honesty)

plot of chunk unnamed-chunk-9

Differences: 1. people don’t think it’s less nice or more mean to say “good”" for 5-heart state compared to 4-heart state 2. saying “okay” is perceived to be less nice and more mean given state of 3, compared to an “okay” true state

Comparing expts: honesty

plot of chunk unnamed-chunk-11plot of chunk unnamed-chunk-11plot of chunk unnamed-chunk-11plot of chunk unnamed-chunk-11

Honesty attribution with word-state looks more comparable to semantic distribution data?