國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Chapter Three: Results
Experiment 1a: Probabilistic risky choice in different EV conditions
The results of Experiment 1a are presented in Figure1, Figure2, and Figure 3, respectively covering the data collected when the EV set on 1, 0.5, and 2 on the PHR arm. In Figure 1, the results of a two-way ANOVA showed a significant main effect of reward ratio (F(2, 10) = 14.613, p < 0.01) and a significant reward-ratio-by-day interaction (F(18, 90) = 2.679, p < 0.01). A test of simple main effect revealed a significant difference of reward ratio factor on the last nine days (from day 2 to 9, in the orders of F(2, 4) = 8.133, 31.778, 21.128, 28.299, 29.000, 22.991, 31.452, 30.117, and 45.408, p < 0.05). Further post hoc comparisons revealed that the percentage of choosing PHR significantly higher in reward ratio 1:2 than reward ratio 1:4 from day 2 to 10 (p < 0.05) and reward ratio 1:8 from day 2 to 10 (p < 0.05) but no difference between reward 1:4 and 1:8 in each of all nine days (p > 0.05).
--- Figure 1
---
In Figure 2, the results of ANOVA showed a significant main effect of reward ratio (F(2, 10) = 6.988, p < 0.05), a significant main effect of day (F(9, 45) = 35.66, p
< 0.001), and a significant reward-ratio-by-day interaction (F(18, 90) = 3.592, p <
0.001). A test of simple main effect revealed a significant difference of reward ratio factor only on the second day (F(2, 4) = 7.484, p < 0.05). Further post hoc
comparisons revealed the percentage of choosing PHR significantly higher in reward ratio 1:4 than reward ratio 1:2 (p < 0.05) and significantly higher in reward ratio 1:8 than reward 1:2 (p < 0.05).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
--- Figure 2
---
In Figure 3, the results of ANOVA showed a significant main effect of reward ratio (F(2, 10) = 8.406, p < 0.01) and a significant main effect of day (F(9, 45) = 21.836, p < 0.001). Post hoc comparisons of reward ratio factor revealed the
percentage of choosing PHR significantly higher in reward ratio 1:2 than reward ratio 1:8 (p < 0.01) but no differences between reward ratio 1:2 and 1:4 neither reward ratio 1:4 and 1:8 (p > 0.05). And post hoc comparisons of day factor revealed the
percentage of choosing PHR significant lower in day 1 than day 2 to 10 (p < 0.01), day 2 lower than day 4 to 10 (p < 0.01), and day 3 lower than day 4 and day 5 (p <
0.05).
--- Figure 3
---
Figure 4 shows the mean values of choosing PHR over the last five days of free choice phase on each of three reward ratios. The results of a two-way ANOVA revealed a significant main effect of EV (F(2, 10) = 77.406, p < 0.001), a significant main effect of reward ratio (F(2, 10) = 15.406, p < 0.01), and a significant
EV-by-reward-ratio interaction (F(4, 20) = 9.214, p < 0.001). A test of simple main effect revealed a significant difference of EV factor on reward ratio 1:2 (F(2, 4) = 2688.648, p < 0.001) and 1:4 (F(2, 4) = 1607.175, p < 0.001) and a significant difference of reward ratio on EV=1 (F(2, 4) = 35.664, p < 0.01). Further post hoc comparisons of EV factor revealed that in reward ratio 1:2, the percentage of choosing
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
PHR significant higher in EV=2 than EV=0.5 (p < 0.001) and EV=1 higher than EV=0.5 (p < 0.001); and in reward ratio 1:4, EV=2 was significantly higher than EV=1 (p < 0.01) and EV= 0.5 (p < 0.001). The post hoc comparisons of reward ratio factor revealed that in EV=1, the percentage of choosing PHR significantly higher in reward ratio 1:2 than reward ratio 1:4 (p < 0.01) and 1:8 (p < 0.01).
--- Figure 4
---
Experiment 1b: Effects of amphetamine on probabilistic risky choice behavior The results of Experiment 1b are presented, based on the condition of EV set in PHR arm as 1, 0.5 and 2, in Figure5, Figure 6, and Figure 7 respectively. In each of these three figures, only the data collected in the last four days are relevant to the drug treatment.
--- Figure 5
--- ---
Figure 6
--- ---
Figure 7
---
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The effect of amphetamine on the present probabilistic risky choice behavior is shown in Figure 8. For the condition of EV=1 set in both PHR and CLR arm, as shown in the top panel of Figure 8, the main effects of dose and reward ratio, F(2, 10)
= 5.798 and F(2, 10) = 6.883 respectively, are significant (p < 0.05). Post hoc comparisons of main effect of dose revealed that the higher dose of amphetamine (1 mg/kg) significantly increased the percentage of choosing PHR compared to saline treatment (p < 0.01). Notably, the effect of lower dose amphetamine (0.5 mg/kg) was only marginal significantly increase the percentage of choosing PHR compared to saline control group (p = 0.051). The post hoc comparisons of main effect of reward ratio revealed the percentage of choosing PHR significantly higher in reward ratio 1:2 than reward ratio 1:4 (p < 0.01) and 1:8 (p < 0.01).
--- Figure 8
---
The data analyses for those shown in the intermediate panel, as revealed by a two-way ANOVA, neither the main effect of dose nor the main effect of reward ratio were significant (p > 0.05). Also, the test of EV-by-reward-ratio interaction was not significant (p > 0.05). For the data presented in the bottom panel of Figure 8, none of the tests of two-way ANOVA was significantly detected (p > 0.05, see Table 1 of Appendix for the details).
--- Table 1
---
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Experiment 2: Effects of lesion manipulations on probabilistic risky choice Histology
The results of histological analyses of bilateral lesions in NAC, DLS, OFC, and mPFC were presented in Figure 9, Figure 10, Figure 11, and Figure 12 respectively.
These rats showed extensive cell collapse and gliosis in the location of the lesion sites.
In Figure 9, the diagram shows the histological examination of NAC. The top panel shows the diagram of the extension of NAC lesions (n = 9). The black area represents the most intensive of lesions area made for all the subjects, whereas the grey area represents the maximum lesion area. The sketch diagram was adopted from the Figure 23 of the rat brain atlas by Paxinos and Watson (2007). Bottom panel shows the representative photographs of coronal sections, and the red circle remarked areas highlight the location of NAC lesions (right photo) and sham lesion (left photo) respectively. The areas of damage in the ventral part of striatum and did not extended to the lateral ventricle.
--- Figure 9
---
In Figure 10, the diagram shows the histological examination of DLS. The top panel shows the diagram of the extension of DLS lesions (n = 9). The black area represents the most intensive of lesions area made for all the subjects, whereas the grey area represents the maximum lesion area. The sketch diagram was adopted from the Figure 27 of the rat brain atlas by Paxinos and Watson (2007). Bottom panel shows the representative photographs of coronal sections, and the red circle
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
remarked areas highlight the location of DLS lesions (right photo) and sham lesion (left photo) respectively.
--- Figure 10
---
In Figure 11, the diagram shows the histological examination of OFC. The top panel shows the diagram of the extension of OFC lesions (n = 9). The black area represents the most intensive of lesions area made for all the subjects, whereas the grey area represents the maximum lesion area. The sketch diagram was adopted from the Figure 9 of the rat brain atlas by Paxinos and Watson (2007). Bottom panel shows the representative photographs of coronal sections, and the red circle remarked areas highlight the location of OFC lesions (right photo) and sham lesion (left photo) respectively.
--- Figure 11
---
In Figure 12, the diagram shows the histological examination of mPFC. The top panel shows the diagram of the extension of mPFC lesions (n = 9). The black area represents the most intensive of lesions area made for all the subjects, whereas the grey area represents the maximum lesion area. The sketch diagram was adopted from the Figure 10 of the rat brain atlas by Paxinos and Watson (2007). Bottom panel shows the representative photographs of coronal sections, and the red circle remarked areas highlight the location of NAC lesions (right photo) and sham lesion (left photo) respectively.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
--- Figure 12
--- Effects of NAC lesion
The results of post-lesion tests on locomotor activity and discrimination test are shown in Figure 13. In the top panel of Figure 13, the locomotor activity of the NAC lesion group was significantly higher than that of the sham lesion control group (t(11) = 2.94, p < 0.05). As the intermediate panel and bottom panel of Figure 13 shows, the results of independent t-test applied on the two measures of the
discrimination task showed no significant difference between the NAC lesion and the sham control groups (p > 0.05).
--- Figure 13
---
Figure 14 shows the effects of NAC lesion on probabilistic risky choice behavior.
The results of a three-way ANOVA revealed a significant main effect of lesion (F(1, 16) = 8.954, p < 0.01), a significant main effect of reward ratio (F(2, 32) = 20.568, p
< 0.001), and a significant main effect of day (F(9, 144) = 28.026, p < 0.001). Also, for the two-way interaction tests, there was a significant reward-ratio-by-day
interaction (F(18, 288) = 10.023, p < 0.001) and a significant lesion-by-day interaction (F(9, 144) = 2.441, p < 0.05). However, the
lesion-by-reward-ratio-by-day interaction was not significant (F(18, 288) = 1.242, p >
0.05). In Figure 15, the tests of simple main effect of lesion-by-day interaction revealed significantly difference in lesion factor, further post hoc revealed a decrease
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
in NAC lesion group on day 4, day 5, day 7, day 8, day 9, and day 10 (p < 0.05).
The of day factor in lesion-by-day interaction revealed significant difference in NAC lesion group (F(9, 8) = 5.701, p < 0.05). Further post hoc comparisons revealed that the percentage of choosing PHR significantly higher in day 1 than day 2 to day 9 (p <
0.05), day 2 higher than day 3 to day 10 (p < 0.001), day 3 higher than day 4 to day 10 (p < 0.01), day 4 higher than day 7 to day 8 (p < 0.05), day 5 higher than day 7 to day 8 (p < 0.05), day 6 higher than day 8 and day 9 (p < 0.05), and day 7 higher than day 10 (p < 0.05). Also, a test of simple main effect of reward-ratio-by-day interaction revealed a significant difference of reward ratio factor on each of all ten days (in the orders of F(2, 15) = 12.462, 6.051, 14.865, 10.875, 14.092, 14.042, 21.581, 19.108, 14.510, and 27.064, p < 0.05) and a significant difference of day factor on reward ratio 1:4 (F(9, 8) = 5.364, p < 0.05) and 1:8 (F(9, 8) = 44.287, p < 0.001). Further post hoc comparisons of reward ratio factor revealed the percentage of choosing PHR
was significantly higher in reward ratio 1:2 than reward ratio 1:4 from day 2 to 10 (p
< 0.05) and higher than reward ratio 1:8 in the first 2 days (p < 0.05). Also, reward ratio 1:8 was significantly higher than reward ratio 1:4 in the first 2 days (p < 0.05).
The post hoc comparisons of day factor revealed that in reward ratio 1:4, the
percentage of choosing PHR significantly higher in day 1 than other 9 days (p < 0.05), day 2 higher than day 3 and day 5 to 10 (p < 0.05), day 3 higher than day 10 (p <
0.05), day 5 higher than day 10 (p < 0.05), day 6 higher than day 8 (p < 0.05), day 7 higher than day 8 and day 10 (p < 0.05), and day 9 higher than day 10 (p < 0.05).
And, in reward ratio 1:8, the percentage of choosing PHR significantly higher in day 1 than day 2 to 10 (p < 0.001), day 2 higher than day 3 to 10 (p < 0.05), day 4 higher than day 5 to 10, (p < 0.05), day 5 higher than day 7 and day 10 (p < 0.05), and day 8 higher than day 10 (p < 0.05).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
--- Figure 14
--- ---
Figure 15
---
Figure 16 shows the effects of NAC lesion on probabilistic risky choice behavior in the first 3 days, the intermediate 4 days, and the last 3 days of ten-day free choice test. The main effect of reward ratio was significant in all three blocks (p < 0.01) and would not be mentioned below. In the first 3 days (top panel), there was no significant main effect of lesion (F(1, 16) = 0.2, p > 0.05) nor significant
lesion-by-reward-ratio interaction (F(2, 32) = 0.34, p > 0.05). The results of a two-way ANOVA revealed a significant main effect of lesion (F(1, 16) = 8.792, p <
0.01) in the intermediate 4 days and a significant main effect of lesion (F(1, 16) = 6.939, p < 0.05) in the last 3 days (See Table 2 of Appendix for the details).
--- Figure 16
--- ---
Table 2
--- Effects of DLS lesion
The effects of DLS lesion on locomotor activity and discrimination test are shown in Figure 17. In the top panel of Figure 17, there were no differences on
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
locomotor activity between DLS lesion group and sham lesion group (t(10) = 0.349, p
> 0.05). Also, as lower two panels of Figure 17, there were no differences on the two measures of the discrimination between two groups (p > 0.05).
--- Figure 17
---
Figure 18 shows the effects of DLS lesion on probabilistic risky choice behavior.
The results of a three-way ANOVA revealed a significant main effect of reward ratio (F(2, 32) = 24.687, p < 0.001), and a significant main effect of day (F(9, 144) = 18.684, p < 0.001). Also, for the two-way interaction tests, there was a significant reward-ratio-by-day interaction (F(18, 288) = 18.487, p < 0.001). However, the lesion-by-reward-ratio-by-day interaction was not significant (F(18, 288) = 0.904, p >
0.05). A test of simple main effect of reward-ratio-by-day interaction revealed a significant difference of reward ratio factor on each of last nine days (from day 2 to 9, in the orders of F(2, 15) = 9.369, 15.947, 12.846, 21.426, 25.833, 45.589, 51.143, 65.990, and 71.946, p < 0.01) and significant difference of day factor on reward ratio 1:4 (F(9, 8) = 3.988, p < 0.05) and 1:8 (F(9, 8) = 8.093, p < 0.01). Further post hoc comparisons of reward ratio factor revealed the percentage of choosing PHR was significantly higher in reward ratio 1:2 than reward ratio 1:4 from day 2 to 10 (p <
0.05) and higher than reward ratio 1:8 from day 3 to 10 (p < 0.05). Also, reward ratio 1:8 was significantly higher than reward ratio 1:4 in the first 2 days (p < 0.05).
The post hoc comparisons of day factor revealed that in reward ratio 1:4, the
percentage of choosing PHR significantly higher in day 1 than other 9 days (p < 0.05), day 2 higher than day 3 and day 5 to 10 (p < 0.05), day 3 higher than day 4 (p < 0.05),
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
day 6 higher than day 8 (p < 0.05), and day 7 higher than day 8 (p < 0.05). And, in reward ratio 1:8, the percentage of choosing PHR significantly higher in day 1 than day 2 to 10 (p < 0.05), day 2 higher than day 3 to 10 (p < 0.001), day 3 higher than day 5 to 10, (p < 0.01), day 4 higher than day 6 to 10 (p < 0.05), day 5 higher than day 7 and day 10 (p < 0.05), day 6 higher than day 7 to 10 (p < 0.05), and day 7 higher than day 10 (p < 0.05).
--- Figure 18
---
Figure 19 shows the effects of DLS lesion on probabilistic risky choice behavior in the three blocks of the first 3 days, the intermediate 4 days, and the last 3 days of ten-day free choice test. Only the main effect of reward ratio was significant in all three blocks (p < 0.01). None of the lesion effect or the lesion-by-reward-ratio interaction was significant in all three blocks (p > 0.05, see Table 3 of Appendix for the details).
--- Figure 19
--- ---
Table 3
--- Effects of OFC lesion
The effects of OFC lesion on locomotor activity and discrimination test are shown in Figure 20. In the top panel of Figure 20, the results of independent t-test
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
revealed a significant increase on locomotor activity in OFC lesion group than that of sham lesion group (t(16) = 2.827, p < 0.05). As for the discrimination ability, as the lower two panels of Figure 20, no differences on both of two measures of the
discrimination between two groups were detected (p > 0.05).
--- Figure 20
---
Figure 21 shows the effects of OFC lesion on probabilistic risky choice behavior.
The results of a three-way ANOVA revealed a significant main effect of lesion (F(1, 16) = 9.725, p < 0.01), a significant main effect of reward ratio (F(2, 32) = 17.831, p
< 0.001), and a significant main effect of day (F(9, 144) = 10.257, p < 0.001). Also, for the two-way interaction tests, there was only a significant reward-ratio-by-day interaction (F(18, 288) = 8.437, p < 0.001). The lesion-by-reward-ratio-by-day interaction was not significant (F(18, 288) = 1.322, p > 0.05). A test of simple main effect of reward-ratio-by-day interaction revealed a significant difference of reward ratio factor on last nine days (from day 2 to 9, in the orders of F(2, 15) = 6.952, 11.594, 9.539, 10.971, 16.946, 19.050, 26.919, 49.429, and 34.195, p < 0.01) and significant difference of day factor in reward ratio 1:8 (F(9, 8) = 6.042, p < 0.01).
Further post hoc comparisons of reward ratio factor revealed the percentage of choosing PHR was significantly higher in reward ratio 1:2 than reward ratio 1:4 from day 2 to 10 (p < 0.05) and higher than reward ratio 1:8 from day 4 to 10 (p < 0.05).
Also, reward ratio 1:8 was significantly higher than reward ratio 1:4 in day 2 and day 3 (p < 0.05). The post hoc comparisons of day factor revealed that in reward ratio 1:8, the percentage of choosing PHR significantly higher in day 1 than day 2 to 10 (p
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
< 0.05), day 2 higher than day 3 to 10 (p < 0.05), day 3 higher than day 5 to 10 (p <
0.05), day 4 higher than day 7 to 10 (p < 0.05), day 5 higher than day 8 to 10 (p <
0.05), day 6 higher than day 7 to 10 (p < 0.05), day 7 higher than day 8 to 10 (p <
0.05), and day 8 higher than day 10 (p < 0.05).
--- Figure 21
---
Figure 22 shows the effects of OFC lesion on probabilistic risky choice behavior in the three blocks of the first 3 days, the intermediate 4 days, and the last 3 days of ten-day free choice test. The main effect of reward ratio was significant in all three blocks (p < 0.01). The results of two-way ANOVA revealed only a significant main effect of lesion in the intermediate 4 days (F(1, 16) = 5.592, p < 0.05). None of two-way interaction was significant (p > 0.05, see Table 4 of Appendix for the details).
--- Figure 22
--- ---
Table 4
--- Effects of mPFC lesion
In the sham lesion group, one rat died during surgery, and the other one rat was excluded from analyses of probabilistic risky choice behavior due to persistent of choosing left arm of the T-maze (> 93%) across all three reward ratios.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Figure 23 shows the effects of mPFC lesion on locomotor activity and discrimination test. In the top panel of Figure 23, there were no difference on locomotor activity between mPFC lesion group and sham lesion group (t(15) = 0.539, p > 0.05). Also, as lower two panels of Figure 23, there were no differences on both of two measures of the discrimination between two groups (p > 0.05).
--- Figure 23
---
Figure 24 shows the effects of mPFC lesion on probabilistic risky choice behavior. The results of a three-way ANOVA revealed a significant main effect of reward ratio (F(2, 28) = 7.331, p < 0.01), and a significant main effect of day (F(9, 126) = 8.055, p < 0.001). Also, for the two-way interaction tests, there was a significant reward-ratio-by-day interaction (F(18, 252) = 13.656, p < 0.001).
However, the lesion-by-reward-ratio-by-day interaction was not significant (F(18, 252)
= 1.062, p > 0.05). A test of simple main effect of reward-ratio-by-day interaction revealed a significant difference of reward ratio factor on day 1 and day 4 to 10 (F(2, 15) = 4.001 in day 1; from day 4 to 10, in the orders of F(2, 15) = 6.254, 7.842, 7.386, 13.158, 17.914, 37.756, and 27.067, p < 0.05) and significant difference of day factor in reward ratio 1:8 (F(9, 6) = 37.610, p < 0.001). Further post hoc comparisons of reward ratio factor revealed the percentage of choosing PHR was significantly higher in reward ratio 1:2 than reward ratio 1:4 from day 4 to 10 (p < 0.05) and higher than reward ratio 1:8 in day 1 and day 4 to 10 (p < 0.05). But there was no significant difference between reward ratio 1:4 and 1:8 in day 1 and day 4 to 10 (p > 0.05). The post hoc comparisons of day factor revealed that in reward ratio 1:8, the percentage of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
choosing PHR significantly higher in day 1 than day 2 to 10 (p < 0.05), day 2 higher than day 3 to 10 (p < 0.05), day 3 higher than day 4 to 10 (p < 0.01), day 4 higher than day 7 to 10 (p < 0.05), day 5 higher than day 7 to 10 (p < 0.01), day 6 higher than day 7 to 10 (p < 0.05), day 7 higher than day 10 (p < 0.05), and day 9 higher than day 10 (p < 0.05).
--- Figure 24
---
Figure 25 shows the effects of mPFC lesion on probabilistic risky choice
behavior in the three blocks of the first 3 days, the intermediate 4 days, and the last 3 days of ten-day free choice test. In the first 3 days, a two-way ANOVA revealed only a significant lesion-by-reward-ratio interaction (F(2, 28) = 3.755, p < 0.05). In the intermediate 4 days and last 3 days, the main effect of reward ratio was significant, F(2, 28) = 11.509, F(2, 28) = 17.693 respectively (p < 0.001, see Table 5 of Appendix
for the details).
--- Figure 25
--- ---
Table 5
---