Table 1.

Task sequence in each cohort

Lighting condition	Cohort	Recognition task sequence
Condition 1: LD	1A (n = 8)	H × 6 → (1) object × 4 → (2) object displacement × 2 → (3) object-in-place × 2
	1B (n = 8)	H × 6 → (1) object × 4 → (2) object displacement × 2 → (3) object-in-place × 2 → (4) odor × 4
	1Ca (n = 8)	H × 6 → (1) object × 4 → (2) object displacement × 2 → (3) object-in-place × 2
Condition 2: rLL	2Ab (n = 8)	H × 6 → (1) object displacement × 2 → (2) object-in-place × 2 → (3) odor × 4
	2Bc (n = 8)	H × 6 → (1) odor × 4 → (2) object displacement × 2 → (3) object-in-place × 2 → (4) object × 4
	2Cb (n = 8)	H × 6 → (1) object displacement × 2 → (2) odor × 2
Control LL	3A (n = 16)	H × 6 → (1) object × 2
	3B (n = 8)	H × 6 → (1) object-in-place × 2
	3C (n = 16)	H × 6 → (1) object × 2 (LD) → (2) odor × 2 (LL)

All mice received 6 habituation trials (H) in the empty arena (3 trials at ZT 6 and 3 trials at ZT 18 under LD) before recognition testing. In each cohort, half of the recognition trials were given at ZT/CT 6 and the remaining trials were given at ZT/CT 18.

^aSample–test delay was 24 h for cohort 1C; for all other cohorts, the sample–test delay was 5 min.

^bCohorts 2A and 2C experienced one cycle of LD–LL alternation before recognition testing.

^cCohort 2B experienced four cycles of LD–LL alternation before recognition testing.