Research reportSustained performance by common marmosets in a delayed matching to position task with variable stimulus presentations

- Common marmosets were trained in a delayed matching to position task.
- Stimulus positions were varied in order to avoid memorization by body position.
- Accurate performance was maintained even after 100 s delay intervals.
- The response times suggested differing strategies depending on the delay length.
- The results supported the robust positional memory of marmosets in the wild.

Working memory is used to solve various cognitive problems by maintaining information for some time and then by refreshing this information after certain purposes are achieved. In the present study, we explored the ability of common marmosets to perform a delayed matching to position (DMTP) task in a controlled environment using operant conditioning. The DMTP task requires the subjects to respond to the sample stimulus and to select one of two comparison stimuli with a position matching that of the sample stimulus after a programmed delay period. Positional arrangement of the sample and comparison stimuli, which were quasi-randomly determined in each trial, was employed to prevent the subjects from using any strategies based on their own body positions or orientations. The delay intervals between presentations of the sample and comparison stimuli were fixed at 0.5 and 1 s in the initial phases and were then varied between 5 intervals per delay set (e.g., 0.5, 1, 2, 4, and 8 s) intermixed in a session. The longest delay interval within a set was gradually increased after the marmosets achieved the criterion of each task. The subjects were successfully trained in the procedure and showed accurate performance even following delays of more than 100 s. The response times in the trials suggested that they used different strategies depending on the delay interval length. Thus, the present study shows the robust ability of common marmosets in a task requiring positional memory, which is related to their foraging strategy observed in the wild.