Article Figures & Data
Figures
- Figure 1.
Outline of the spatial self-ordered sequencing task showing the experimental timeline (A), touchscreen stimuli positions (B), and variable and fixed array tasks (C,D). An example of a correct sequence is shown (E), as well as perseverative errors of the recurrent (F) and continuous (G) type. Details relating to counterbalancing can be found in Extended Data Table 1-1A–F.
- Figure 2.
Cannulae placements for each of the six experimental marmosets located within the vlPFC projection zone in the caudate and putamen. Males are open symbols, females filled symbols. Square, M1 and M2; triangle, M3 and M4; inverted triangle, M5 and M6.
- Figure 3.
Task performance accuracy on variable and fixed array tasks (A–C) and entropy measures across training compared with saline infusion performance on the fixed task (D). Arrow on the y-axis indicates the entropy score if correct sequences were selected randomly (2.58 bits). For the variable array (A) one-way rmANOVA revealed a main effect of training stage on accuracy (F(2,10) = 20.45; p = 0.0003); Sidak-corrected post hoc showed the following: Training Stage 1 versus Final Performance (p = 0.0032) and Training Stage 2 versus Final Performance (p = 0.0003). For the fixed array (B) one-way rmANOVA showed a main effect of training stage on accuracy (F(2,10) = 8.512; p = 0.0069); Sidak-corrected post hoc showed the following: Selected Sequence versus Final Performance (p = 0.0099) and Training Stage 1 versus Final Performance (p = 0.0263). The significant difference between variable and fixed array accuracy (C) was determined via paired t test (p = 0.0014). Finally, a one-way rmANOVA showed a main effect of training stage on entropy (F(2,10) = 20.81; p = 0.0003); Sidak-corrected post hoc showed the following: Selected Sequence versus Initial Training (p = 0.0044) and Selected Sequence versus Final Performance (p = 0.0003). Data are represented as mean ± standard error of mean and Sidak comparisons indicated by *p > 0.05, **p > 0.01, and ***p > 0.001; ns, not significant. Data are represented as mean ± standard error of mean. Data were collected from six monkeys, each with their own symbol as designated in Figure 2.
- Figure 4.
Effects of 1 or 3 mM CNQX into either the caudate nucleus, the putamen, or combined caudate and putamen on variable (A) and fixed (B) array accuracy. Data were collected across all regions and both CNQX doses from six monkeys, each with their own symbol defined in the key. For the variable array (A) two-way rmANOVA revealed a main effect of treatment (F(2,10) = 13.04; p = 0.0016); but no effect of brain region (F < 1). Sidak-corrected post hoc showed the following: caudate saline versus 3 mM CNQX (p = 0.0088), putamen saline versus 3 mM CNQX (p = 0.0073), combined caudate + putamen saline versus 1 or 3 mM CNQX (p = 0.0064; p < 0.0001, respectively). The fixed array (B), two-way rmANOVA revealed a main effect of treatment (F(2,10) = 23.91; p = 0.0002) but no effect of brain region F(2,10) = 2.495; p = 0.13211); Sidak-corrected post hoc showed the following: caudate saline versus 3 mM CNQX (p = 0.265), putamen saline versus 3 mM CNQX (p = 0.0007), and combined caudate + putamen saline versus 3 mM CNQX (p = 0.0033). Data are represented as mean ± standard error of mean and Sidak comparisons indicated by *p > 0.05, **p > 0.01, ***p > 0.001; or ***p > 0.0001. Additional control data for the variable and fixed arrays are presented in Extended Data Figure 4-1 and Extended Data Figure 4-2, respectively.
- Figure 5.
Effects of 1 or 3 mM CNQX into either the caudate nucleus, the putamen, or combined caudate and putamen on perseverative errors of the continuous (A,C) and recurrent type (B,D) in variable (A,B) and fixed (C,D) arrays. Data were collected across all timepoints from six monkeys, each with their own symbol defined in the key. For the variable array, two-way rmANOVAs revealed a main effect of treatment on continuous (F(2,10) = 4.694; p = 0.0365) and recurrent errors (F(2,10) = 7.287; p = 0.0112) but no effect on brain region (F < 1 for both continuous and recurrent errors). Sidak-corrected post hoc showed for variable array continuous errors the following: putamen saline versus 3 mM CNQX (p = 0.0011) and combined caudate + putamen saline versus 1 mM or 3 mM CNQX (p = 0.0164; p = 0.0004, respectively). For variable array recurrent errors, it revealed the following: caudate saline versus 3 mM CNQX (p = 0.0701) and combined caudate + putamen saline versus 3 mM CNQX (p = 0.0755). The fixed array two-way rmANOVAs revealed a main effect of treatment on both continuous (F(2,10) = 6.481; p = 0.0157) and recurrent errors (F(2,10) = 15.56; p = 0.0009) but no effect of brain region (F(2,10) = 2.477; p = 0.1337 for continuous errors, and F < 1 for recurrent errors). Sidak-corrected post hoc showed for fixed array continuous errors as follows: caudate saline versus 3 mM CNQX (p = 0.0429) and putamen saline versus 3 mM CNQX (p = 0.0029). For fixed array recurrent errors, it revealed: putamen saline versus 3 mM CNQX (p = 0.0069) and combined caudate + putamen saline versus 3 mM CNQX (p = 0.0062). Data are represented as mean ± standard error of mean and Sidak comparisons indicated by *p > 0.05, **p > 0.01, or ***p > 0.001.
Tables
1 mM CNQX 3 mM CNQX Caudate Putamen Caudate + Putamen Caudate Putamen Caudate + Putamen Variable array Accuracy - - Impaired Impaired Impaired Impaired Continuous errors - - Increased - Increased Increased Recurrent errors - - - Trending up - Trending up Fixed array Accuracy - - - Impaired Impaired Impaired Continuous errors - - - Increased Increased Trending up Recurrent errors - - - - Increased Increased
Table 1-1
Individual marmoset details and counterbalancing tables showing order of array type, brain region targeted, and concentration of CNQX administered. Download Table 1-1, XLSX file.
Figure 4-1
Typical experimental schedule for monkeys on the variable array (A). Control experiments showing that accuracy (B-D) and error types (E-J) are not affected by day of the week (B, E, H) or the animal being held (C, F, I). In addition, receiving an infusion (saline) or being held for consecutive days (D, G, J) has no effect on Friday performance (when CNQX infusions are performed). Each data point represents an average of three measurements for the weekday baseline measures (B, E, H); an average of 6-10 values for no-handling-mock comparisons (C, F, I), and single values for Mock/Saline/Mock comparisons (D, G, and J). Each symbol on a graph represents an individual monkey. Paired t-tests (n=3) showed that there was no difference in accuracy (p>0.05), or errors of the continuous (p>0.05) or recurrent (p>0.05) kind (B, E, H); nor were there significant effects (p>0.05) on accuracy or either error type when the animal was handled during mocks compared to no handling (C, F, I). A one-way rmANOVA revealed no main effects after being held on consecutive days/receiving a control infusion on accuracy (F(2, 6) <1), or errors of the continuous (F(2, 2) = 10.33; p=0.0882) or recurrent (F(1, 1) = 2.385; p=0.3658) kind (D, G, J). Download Figure 4-1, TIF file.
Figure 4-2
Typical experimental schedule for monkeys on the fixed array (A). Control experiments showing that accuracy (B-D) and error types (E-J) are not affected by day of the week (B, E, H) or the animal being held (C, F, I). In addition, receiving an infusion (saline) or being held for consecutive days (D, G, J) has no effect on Friday performance (when CNQX infusions are performed). Each data point represents an average of three measurements for the weekday baseline measures (B, E, H); an average of 6-10 values for baseline-mock comparisons (C, F, I), and single values for Mock/Saline/Mock graphs (D, G, and J). Each symbol on the graphs indicates a different monkey. A one-way rmANOVA revealed no main effect after being held on consecutive days/receiving a control infusion on accuracy (F (2, 6) = 3-355; p=0.1052), or errors of the continuous (F (2, 2) = 7; p=0.1250) or recurrent (F<1) kind (D, G, J). A one-way rmANOVA revealed no main effect of day of the week on accuracy (F<1), or errors of the continuous (F<1) or recurrent (F<1) kind (B, E, H). A paired t-test showed no significant effects (p>0.05) on accuracy or either error type when the animal was handled during mocks compared to no handling (C, F, I). A one-way rmANOVA revealed no main effect after being held on consecutive days/receiving a control infusion on accuracy (F <1), or errors of the continuous (F <1) or recurrent (F<1) kind (D, G, J). Download Figure 4-2, TIF file.
In this issue
