Table 2 Mean (SD) errors and response latencies and probabilities of win–stay/lose–shift behaviour based on feedback type (80% valid, 20% misleading) in the two treatment groups

From: Dissociable effects of acute SSRI (escitalopram) on executive, learning and emotional functions in healthy humans

 

Measure

 

Placebo

Escitalopram

Group differencea

Stage 1

Errors

 

1.76 (3.73)

4.55 (6.66)

p = 0.004

 

RTs (ms)

 

2016.06 (937.32)

2165.79 (968.95)

p = 0.532

 

Feedback type

80% accurate

20% misleading

    

Probabilities

Reward–stay

80%

86.74

75.6

p = 0.043

  

20%

1.14

4.03

p = 0.104

 

Reward–shift

80%

5.97

9.48

p = 0.125

  

20%

4.92

8.47

p = 0.279

 

Lose–stay

80%

1.42

4.13

p = 0.125

  

20%

88.64

72.18

p = 0.009

 

Lose–shift

80%

5.87

10.79

p = 0.032

  

20%

5.3

15.32

p = 0.003

Stage 2

Errors

 

6.58 (2.57)

7.58 (4.46)

p = 0.689

 

RTs (ms)

 

2251.86 (6285.86)

1280.32 (865.23)

p = 0.237

 

Feedback type

80% accurate

20% misleading

    

Probabilities

Reward–stay

80%

68.75

63.71

p = 0.376

  

20%

11.74

15.32

p = 0.359

 

Reward–shift

80%

6.06

8.57

p = 0.238

  

20%

4.92

6.58

p = 0.442

 

Lose–stay

80%

16.1

16.32

p = 0.237

  

20%

70.08

62.9

p = 0.971

 

Lose–shift

80%

10.07

12.6

p = 0.684

  

20%

13.26

14.92

p = 0.355

  1. aGroup difference: p-values of one-way ANOVAs. Significant p-values following control of the false discovery rate of q < 0.15 with Benjamini–Hochberg procedure are shown in bold