Table 1.
Optimized design and schedule for our visual transitive inference touchscreen task in mice
| Stage | Design | Trials/session | Performance criterion |
|---|---|---|---|
|
Training stage 1: Stimulus exposure |
Exposure to all stimulus pairs to facilitate flexibility: 1 session of each stimulus pair in order: A+B−; B+C−; C+D−; D+E− | 52 trials/session; Total 4 sessions | None |
| Training stage 2: Premise stimulus pair learning | Learning each pairwise discrimination to criterion in order: A+B−; B+C−; C+D−; D+E− | 52 trials/session; until criterion | ≥ 41/52 correct (79% accuracy) in a session, plus one of the two previous or successive sessions at ≥ 34/52 correct (≥ 65% accuracy) |
|
Training stage 3: Serial integration |
All stimulus pairs presented in serial order: A+ B−; B+ C−; C+ D−; D+ E− then sequence repeated with 13 trials of each pair presented within a session | 52 trials/session; until criterion | ≥ 37/52 correct (≥ 70% accuracy) in a session or max. 15 sessions (780 trials) |
| Training stage 4: Pseudorandom integration | All stimulus pairs (A+B−; B+C−; C+D-; D+E−) presented in pseudorandom order with 13 trials of each pair presented within a session | 52 trials/session; until criterion | ≥ 37/52 correct (≥ 70% accuracy) in a session or max. 15 sessions (780 trials) |
| Transitive Inference Test | 20 unrewarded trials each of novel pairs AE and BD; 8 trials each of the learned pairs (A+B−, B+C−, C+D−, D+E−); all pairs presented in pseudorandom order within a session | 72 trials/session; 1 session | None |