Article Figures & Data
- Figure 1.
Diagram of touchscreen training and testing schedules. Top, An image of the touchscreen chamber is shown when configured for pairwise visual discrimination. The stages of training for pairwise visual discrimination (left) and delayed nonmatch to position (right) are shown, with similar pretraining shown in the middle. Pretraining for both tasks consists of autoshaping to the food magazine, FR-1 to the illuminated square/image (without punishment for presses to the blank space), and FR-1 to the illuminated square/image (with a punished timeout period for presses to the blank space). Pairwise visual discrimination pretraining included the following two additional stages: (1) after autoshaping, mice received 1 d where they received reward either for an active screen touch, or after 30 s, whichever came first, and thereafter, all trials were “forced trials”; and (2) before punishment was added for blank responses, mice had to “initiate” trials with a nosepoke into the food magazine. Subsequently, pairwise visual discrimination consists of the discrimination between S+ and S−, followed by reversal of reward contingencies. The images used in the present study are shown (above). For the nonmatch task, after abbreviated pretraining, mice first learned the nonmatch rule without delays, followed by short delays (1 and 3 s), and then were tested for 25 d at the full delay schedule (1, 3, and 10 s delays).
- Figure 2.
Pairwise visual discrimination showed no genotype differences in performance between Fmr1 and WT mice. A, Days to criterion for acquisition and reversal of mice completing both phases. B, Trials to criterion for acquisition and reversal of mice completing both phases. C, Days to criterion for acquisition, indicating proportion of individuals that had completed training at each day (survival curve). D, Days to criterion for reversal, indicating the proportion of individuals that had completed reversal at each training day (survival curve).
- Figure 3.
Validation of the delayed nonmatching to position task revealed superior working memory performance in B6 compared with FVB/AntJ inbred strains of mice. A, B6 performance at 1, 3, and 10 s delays on the delayed nonmatching to position task. B, FVB/AntJ performance at 1, 3, and 10 s delays. C, D, Days to criterion (C) and trials to criterion (D) for nonmatching to position rule learning (without delays), and acquisition of initial delayed nonmatching to position (1 and 3 s delays only). E, Days to criterion (survival curves) for nonmatching rule acquisition, indicating the proportion of individuals that had completed training at each training day. F, Days to criterion for the acquisition at 1 and 3 s delays, indicating the proportion of individuals that had completed this phase at each training day. *p < 0.05 compared with 1 s delay (A, B); *p < 0.05, strain difference for all days and trials to criterion in C–F.
- Figure 4.
Delayed nonmatching to position showed no genotype differences between Fmr1 and WT mice. A, WT performance on delayed nonmatching to position at 1, 3, and 10 s delays. B, Fmr1 performance on delayed nonmatching to position at 1, 3, and 10 s delays. C, D, Days to criterion (C) and trials to criterion (D) for rule learning in nonmatching to position (without delays) and acquisition of the initial delay periods (1 and 3 s delays only). E, Days to criterion for nonmatch rule acquisition, indicating the proportion of individuals that had completed training at each training day. F, Days to criterion for short delay (1 and 3 s) acquisition, indicating the proportion of individuals that had completed this phase across time. *p < 0.05 indicates significant difference compared with 1 s delay on full schedule performance.
- Figure 5.
Acquisition of Morris water maze hidden platform spatial navigation learning showed no genotype differences between Fmr1 and WT mice. Both genotypes displayed normal performance during acquisition. A, Latency to find the hidden platform. B, Distance traveled during the training trials. C, Swim speed during the training trials. D, Quadrant time during the 60 s probe trial, begun 3 h after the last training trial. E, Platform location crossings during the 60 s probe trial. *p < 0.05 indicates more time in the previously trained platform quadrant than in the other three quadrants, and more crossings over the previous platform location than over the other three pseudolocations.
- Figure 6.
Reversal of Morris water maze hidden platform spatial navigation learning showed no genotype differences between Fmr1 and WT mice. Both genotypes displayed normal performance during reversal. A, Latency to find the hidden platform. B, Distance traveled during the training trials. C, Swim speed during the training trials. D, Quadrant time during the 60 s probe trial, begun 3 h after the last training trial. E, Platform location crossings during the 60 s probe trial. *p < 0.05 indicates more time in the previously trained platform quadrant than in the other three quadrants, or more crossings over the previous platform location than over the other three pseudolocations.
- Table 1:
Statistical results for pairwise visual discrimination acquisition and reversal in Fmr1 and WT mice
Effect Data structure Type of test Power df (between) df (within) F χ2 p Pairwise discrimination phase (d) Normally distributed Two-factor repeated-measures ANOVA 0.96 1 9 17.58 0.002 Pairwise discrimination genotype (d) Normally distributed Two-factor repeated-measures ANOVA 0.41 1 9 3.78 0.08 Pairwise discrimination interaction (d) Normally distributed Two-factor repeated-measures ANOVA with post hoc Bonferroni correction 0.13 1 9 0.83 0.4 Pairwise discrimination phase (trials) Normally distributed Two-factor repeated-measures ANOVA 0.62 1 9 6.45 0.03 Pairwise discrimination genotype (trials) Normally distributed Two-factor repeated-measures ANOVA 0.16 1 9 1.10 0.3 Pairwise discrimination interaction (trials) Normally distributed Two-factor repeated-measures ANOVA 0.06 1 9 0.13 0.7 Pairwise discrimination (survival curve) Normally distributed Mantel–Cox test 0.06 1 0.53 0.5 Pairwise discrimination reversal (survival curve) Normally distributed Mantel–Cox test 0.07 1 0.48 0.5 - Table 2:
Statistical results for nonmatching to position and delayed nonmatching to position performance in B6 and FVB mice
Effect Data structure Type of test Power df (between) df (within) F χ2 t p B6 day (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 0.78 24 216 0.97 0.5 By delay (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 1.00 2 18 386.96 1.67E-15 B6 interaction (DNMTP) Sphericity passed Two-factorrepeated-measures ANOVA with post hoc Bonferroni correction 0.94 48 432 0.92 0.6 Simple effects test (DNMTP) Normally distributed One-way ANOVA with post hoc Bonferroni correction 1.00 2 18 387.00 1.67E-15 1 vs 3 s (DNMTP) Normally distributed Post hoc Bonferroni correction 1.00 9 4.99 0.0004 3 vs 10 s (DNMTP) Normally distributed Post hoc Bonferroni correction 1.00 9 21.21 0.00000002 FVB day (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 0.73 24 216 0.89 0.6 FVB delay (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 1.00 2 18 582.60 3.22E-15 FVB interaction (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA with post hoc Bonferroni correction 0.98 48 432 1.10 0.3 Simple effects test (DNMTP) Normally distributed One-way ANOVA with post hoc Bonferroni correction 1.00 2 18 582.60 3.22E-15 1 vs 3 s Normally distributed Post hoc Bonferroni correction 1.00 9 13.02 0.0000002 3 vs 10 s Normally distributed Post hoc Bonferroni correction 1.00 9 20.81 0.000000007 Nonmatch acquisition (strain survival curve) Not normally distributed Mantel–Cox test 0.11 1 0.03 0.9 Initial delay acquisition (strain survival curve) Not normally distributed Mantel–Cox test 0.73 1 7.83 0.005 Nonmatch and delayed nonmatch acquisition (strain) Normally distributed Two-factor repeated-measures ANOVA 0.80 1 18 8.70 0.009 Nonmatch and delayed nonmatch acquisition (phase) Normally distributed Two-factor repeated-measures ANOVA 0.35 1 18 2.72 0.1 Nonmatch and delayed nonmatch acquisition (interaction) Normally distributed Two-factor repeated-measures ANOVA with post hoc Bonferroni correction 0.60 1 18 5.41 0.03 Strain comparison (strain) Normally distributed Mixed-model ANOVA 0.05 1 18 0.002 0.97 Strain comparison (delay) Normally distributed Mixed-model ANOVA 1.00 2 36 869.6 0.000000 Strain comparison (interaction) Normally distributed Mixed-model ANOVA 1.00 2 36 20.68 0.000001 - Table 3:
Statistical results for nonmatching to position and delayed nonmatching to position in Fmr1 and WT mice
Effect Data structure Type of test Power df (between) df (within) F χ2 t p WT day (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 0.92 24 192 1.34 0.1 WT delay (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 1.00 2 16 237.31 1.28E-12 WT interaction (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA with post hoc Bonferroni correction 0.93 48 384 0.90 0.7 Simple effects test (DNMTP) Normally distributed One-way ANOVA with post hoc Bonferroni correction 1.00 2 16 237.30 1.28E-12 1 vs 3 s Normally distributed Post hoc Bonferroni correction 1.00 8 8.25 0.0002 3 vs 10 s Normally distributed Post hoc Bonferroni correction 1.00 8 13.34 0.000007 FMR1 day (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 0.53 24 168 0.65 0.9 FMR1 delay (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA 1.00 2 14 162.69 0.0000000002 FMR1 interaction (DNMTP) Sphericity passed Two-factor repeated-measures ANOVA with post hoc Bonferroni correction 0.96 48 336 1.01 0.5 Simple effects test (DNMTP) Normally distributed One-way ANOVA with post hoc Bonferroni correction 1.00 2 14 162.70 0.0000000002 1 vs 3 s Normally distributed Post hoc Bonferroni correction 1.00 7 6.57 0.0009 3 vs 10 s Normally distributed Post hoc Bonferroni correction 1.00 7 11.26 0.0000003 Nonmatch acquisition (genotype survival curve) Not normally distributed Mantel–Cox test 0.98 1 0.11 0.7 Initial delay acquisition (genotype survival curve) Not normally distributed Mantel–Cox test 0.95 1 0.13 0.7 Nonmatch and delayed nonmatch acquisition (genotype) Normally distributed Two-factor repeated-measures ANOVA 0.14 1 17 0.84 0.4 Nonmatch and delayed nonmatch acquisition (phase) Normally distributed Two-factor repeated-measures ANOVA 0.69 1 17 6.85 0.02 Nonmatch and delayed nonmatch acquisition (interaction) Normally distributed Two-factor repeated-measures ANOVA 0.05 1 17 0.05 0.8 Genotype comparison (genotype) Normally distributed Mixed-model ANOVA 0.09 1 15 0.4272 0.5 Genotype comparison (delay) Normally distributed Mixed-model ANOVA 1.00 2 30 393.9 0.000000 Genotype comparison (interaction) Normally distributed Mixed-model ANOVA 0.28 2 30 1.43 0.3 - Table 4:
Statistical results for Morris water maze (MWM) acquisition performance in Fmr1 and WT mice
Effect Data structure Type of test Power df (between) df (within) F P MWM acquisition latency-genotype Sphericity violated Mixed-model ANOVA 0.19 1 22 1.31 0.3 MWM acquisition latency-day Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 1.00 4.63 101.95 25.73 5.55E-16 MWM acquisition latency-interaction Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 0.26 4.63 101.95 0.61 0.7 MWM acquisition distance-genotype Sphericity passed Mixed-model ANOVA 0.12 1 22 0.62 0.4 MWM acquisition distance-day Sphericity passed Mixed-model ANOVA 1.00 7 154 16.26 6.66E-16 MWM acquisition distance-interaction Sphericity passed Mixed-model ANOVA 0.36 7 154 0.85 0.5 MWM acquisition speed-genotype Sphericity violated Mixed-model ANOVA 0.32 1 22 2.47 0.1 MWM acquisition speed-day Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 1.00 4.49 98.82 9.64 0.0000005 MWM acquisition speed-interaction Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 0.29 4.49 98.82 0.69 0.6 WT quadrant time Sphericity passed Repeated-measures ANOVA with post hoc Dunnett’s test 0.97 3 33 7.19 0.0008 FMR1 quadrant time Sphericity violated Repeated-measures ANOVA with Greenhouse–Geisser correction and post hoc Dunnett’s test 1.00 1.65 18.16 25.04 0.00001 WT platform crossings Sphericity passed Repeated-measures ANOVA with post hoc Dunnett’s test 0.99 3 33 12.54 0.00001 FMR1 platform crossings Sphericity violated Repeated-measures ANOVA with Greenhouse–Geisser correction and post hoc Dunnett’s test 1.00 1.36 14.98 28.16 0.00003 - Table 5:
Statistical results for Morris water maze reversal performance in Fmr1 and WT mice
Effect Data structure Type of test Power df (between) df (within) F p Latency reversal-genotype Sphericity violated Mixed-model ANOVA 0.05 1 22 0.001 0.97 Latency reversal-day Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 1.00 2.16 47.58 10.12 0.0002 Latency reversal-interaction Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 0.22 2.16 47.58 0.82 0.5 Distance reversal-genotype Sphericity violated Mixed-model ANOVA 0.07 1 22 0.16 0.7 Distance reversal-day Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 0.99 2.17 47.66 7.91 0.0008 Distance reversal-interaction Sphericity violated Mixed-model ANOVA with Greenhouse–Geisser correction 0.37 2.17 47.66 1.46 0.2 Speed reversal-genotype Sphericity passed Mixed-model ANOVA 0.05 1 22 0.04 0.8 Speed reversal-day Sphericity passed Mixed-model ANOVA 0.41 3 66 1.63 0.2 Speed reversal-interaction Sphericity passed Mixed-model ANOVA 0.33 3 66 1.28 0.3 WT reversal quadrant time Sphericity violated Repeated-measures ANOVA with Greenhouse–Geisser correction and post hoc Dunnett’s test 0.80 1.94 21.37 4.08 0.3 FMR1 reversal quadrant time Sphericity violated Repeated-measures ANOVA with Greenhouse–Geisser correction and post hoc Dunnett’s test 1.00 2.02 22.24 20.84 0.000008 WT reversal platform crossings Sphericity passed Repeated-measures ANOVA with post hoc Dunnett’s test 0.91 3 33 5.41 0.004 FMR1 reversal platform crossings Sphericity violated Repeated-measures ANOVA with Greenhouse–Geisser correction and post hoc Dunnett’s test 1.00 1.97 21.68 9.91 0.0009
- Movie 1.
Performance in the touchscreen apparatus on delayed nonmatching to position in a representative FVB/AntJ mouse. Two trials are shown, and the sample appears in the left location for both. After touching the sample image, the mouse turns to the back of the chamber to nose poke in the reward tray. After a random delay (1, 3, or 10 s), the nose poke initiates a choice (match and nonmatch) where the image appears on both sides of the touchscreen. For both trials shown, the mouse correctly nonmatches and earns a reward.
In this issue
