PMC

Glucose meter comparisons. (A) Two different glucose meters give different results. A blood sample was collected from individual fish, and each sample (n = 31) was measured immediately with both meters. A broad range of glucose values was compared by measuring fish that had been recently fed (beginning at 15 min postprandial) as well as fish that had been fasted for up to 8 days. (B) Blood glucose measured with pyrroloquinoline quinone glucose dehydrogenase (Meters B–D) using different sample sizes, compared with glucose oxidase (Meter A). Test strip sample volumes: Meter A, 1.0 μL; Meter B, 0.3 μL; Meter C, 0.6 μL; Meter D, 1.5 μL. Samples (n = 10) from individual fish were measured once with each meter. Order of meter use was randomized. Tukey's multiple comparison posttest revealed that Meter B versus Meter C was not statistically significant (p > 0.05). All other comparisons were significant at p < 0.001, except Meter D versus Meter A was significant at p < 0.05. Data are mean and standard deviation.

(A) Correct responses to triple sequences. (B) Correct responses to duple sequences. Open bars: unaccented condition data. Gray bars: accented conditions. Hashed open bars: bimodal conditions. Hashed gray bars: bimodal accented conditions. Error bars are standard error. (C) Discriminability analysis of meter perception with Aud-C1/Tac-C2 condition; (D) Discriminability analysis of meter perception with Tac-C1/Aud-C2 condition. Open bars are unimodal unaccented control condition, hashed bars are bimodal unaccented condition; gray bars are unimodal accented control condition, and hashed gray bars are bimodal accented condition.
doi:10.1371/journal.pone.0048496.g006

(A) Correct responses to triple sequences. (B) Correct responses to duple sequences. Open bars: unaccented condition data. Gray bars: accented conditions. Hashed open bars: bimodal conditions. Hashed gray bars: bimodal accented conditions. Error bars are standard error. (C) Discriminability analysis of meter perception with Aud-C1/Tac-C2 condition; (D) Discriminability analysis of meter perception with Tac-C1/Aud-C2 condition. Open bars are unimodal unaccented control condition, hashed bars are bimodal unaccented condition; gray bars are unimodal accented control condition, and hashed gray bars are bimodal accented condition.

(A) triple sequences tested in the M/N split task, (B) triple sequences tested in the M/N half-split task, (C) discriminability analysis of meter perception in the M/N split and M/N half-split tasks with Aud-C1/Tac-C2 condition, (D) discriminability analysis of meter perception in the M/N split and M/N half-split tasks with Tac-C1/Aud-C2 conditions. Open bars are results tested under unimodal condition. Hashed bars are results tested under bimodal condition. Error bars are standard error.
doi:10.1371/journal.pone.0048496.g005