Answers are grounded in this paper and cite their sources.
One rat was trained to respond to two four-item-series, A1-B1-C1-D1 and A2-B2-C2-D2, concurrently by responding to four barriers of modified Hill maze in a fixed order. Two types of novel series in which items were exchanged between the two original series were inserted to training as test series. In the congruent test series items that had same serial positions were exchanged between series (e.g., A1-B2-C1-D1, A2-B2-C1-D2), whereas in the incongruent test series items of different serial positions were exchanged (e.g., A1-C2-C1-D1, A2-B2-B1-D2). Rats showed reliably better performance to the congruent test series than the incongruent test series. The item association learning cannot explain this result because the same number of new item associations between exchanged items and adjacent items should be required in the congruent and the incongruent series. Therefore present findings strongly suggest that rats can learn serial position of item B and C of a four-item-series in a certain kind of serial learning procedure.
No takes yet. Share an insight, caveat, or question.
Taniuchi et al. (2012) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: