A novel task to assess mood congruent memory bias in non-human animals.

Background: Assessing the affective state of animals is important for a range of research areas, including neuroscience. The use of cognitive judgement and attention biases to determine affective state has been demonstrated in animals, but approaches to assess mood-congruent biases in memory have yet to become established.

New Method: We describe a novel methodology to investigate memory bias in animals, presenting initial data using the influence of social status to manipulate affective state. The method required laboratory rats to achieve criterion at a working-memory task in an eight-arm radial maze before probing their memory of putative negative, positive or neutral events that occurred in specific arms of the maze. They were tested 2 h and 24 h after experiencing each event to determine how the affective valence of the event influenced task performance.

Results: Regardless of social status, rats avoided arms where they had experienced negative events and preferred arms where they had experienced positive events. However, subordinate rats made errors sooner than dominant rats in tests following exposure to the negative event. Furthermore, whilst subordinate individuals made errors earlier in tests following the negative event relative to the neutral or positive event, dominant rats made errors earlier in tests that followed the positive event.

Comparison With Existing Method(s): Changes in performance thus appeared to reflect social status and associated affective state, confirming a new method for assessing animal affect.

Conclusions: This new memory bias task could potentially be used to determine affective state in a range of non-human animal species.