Difference between revisions of "CNP DRL"

From Pheno Wiki
Jump to: navigation, search
Line 1: Line 1:
 +
go back to [[HTAC]]
 +
 +
 +
=== Basic Task Description ===
 +
 +
The deterministic reversal learning task (DRLT) is a computerized measure of reversal learning, a form of behavioral/cognitive flexibility that has been widely investigated in both animal and human studies (Robbins et al., 2000).  This particular task version was modeled after a paradigm originally used to examine genetic associations with reversal learning in BXD strains of mice (Laughlin et al., 2011).  Participants are presented with two doors and instructed that they will receive monetary rewards based on which door they choose.  The initial side of reinforcement (right or left) is randomly selected by the computer.  If participants select the door associated with the reward, they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise).  Rewarded selections also present a running total of earnings on the bottom of the screen.  If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise).  Once participants have consistently selected the reinforced door, the reward contingencies for the doors are reversed.  After this reversal, participants need to consistently select the newly reinforced door in order to complete the task.  The primary dependent variable is the number of errors that are made after reversal.
 +
 +
=== Task Procedure ===
 +
This DRLT was programmed in E-Prime 2.0. Participant responses are made on computer keyboard, with three keys marked for use in this task: A (left), H (middle), and '(right).  At the beginning of the task, instructions are presented on the screen in yellow, size 20 Arial font, with a black screen background. The instructions are as follows:
 +
 +
<center>"In this game, you have an opportunity to win money by
 +
 +
looking behind doors.  Do your best to find as much money as possible.
 +
 +
You will actually be given the money that you earn in this task.
 +
 +
You may only use the index finger of your dominant hand to respond,
 +
 +
and you must respond as quickly as you can.
 +
 +
You may NOT use two hands on this task.
 +
 +
On the keyboard, you'll see three marked keys: left, middle, and right. 
 +
 +
To start each play of the game,
 +
 +
PRESS AND HOLD THE MIDDLE KEY DOWN,
 +
 +
and release the key only when you see doors appear to the left and right. 
 +
 +
Then choose one of the doors to open by pressing either the left or right keys.
 +
 +
You will get money based on which door you choose.
 +
 +
Before we start the game, let's take a moment to practice the key presses..."</center>
 +
 +
Participants are then shown another instruction slide that prepares them for the practice trials:
 +
 +
<center>"In the following trials you will practice making quick key presses.
 +
You'll hear a sound to let you know if your responses are fast enough.
 +
Please tell your experimenter when you are ready to begin the practice trials."</center>
 +
 +
Practice block trials provide a basic orientation to the trial format and response keys in order to shape prompt responding during the task.  Practice trials begin with a box in the center of the screen with the word "Press" and the words "Press and hold the middle button" at the top of the screen.  Once the participant presses the middle key, the word "Hold" appears in the center box and the words "Wait until you see the doors!" appears at the top of the screen.  The "Hold" cue remains on the screen for either 2000 ms or 2500 ms, during which time participants must keep the middle key pressed down.  If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops!  Make sure to wait for the doors. Let's try again."  Following the "Hold" delay, two doors appear on the left and right sides of the screen.  The doors remain on the screen for up to 4000 ms or until the participant presses the left or right key.  While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key.  If a left or right keypress is made within 2000 ms, then participants are given feedback of "Good Job!" in the center box, along with a positive noise (fanfare).  If participants take too long to respond (more than 2000 ms), they receive the following feedback in the center of the screen: "Oops!  Please respond more quickly. Let's try again."  Feedback slides are all presented for a total of 2000 ms.  After three consecutive responses within the 2000 ms time limit, the practice trials end and the following instructions are presented on the screen:
 +
 +
<center>"Now that you are familiar with pressing the keys, we can start the game. 
 +
In this game, you'll need to choose either the left or the right door to win money.
 +
Please tell your experimenter when you are ready to begin."</center>
 +
 +
Experimental block trials follow general parameters similar to the practice trials, and begin with a box in the center of the screen with the word "Press."  If participants do not press the middle button within 5000 ms, they receive the following feedback in the center of the screen: "Please press the middle key to begin the next trial."  After pressing the middle key, the word "Hold" appears in the center box and remains on the screen for either 1000 ms, 1500 ms, 2000 ms, or 2500 ms.  These 4 durations for the "Hold" cue are randomized across all trials.  If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops!  Make sure to wait for the doors. Let's try again."  Following the "Hold" delay, two doors appear on the left and right sides of the screen.  The doors remain on the screen for up to 2000 ms or until the participant presses the left or right key.  While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key. 
 +
 +
For each experimental block of trials, the computer randomly selects either the left or right door to be initially associated with reward.  Participants receive $0.50 each time they select the rewarded door.  As soon as the participant presses the key corresponding to the side of the rewarded door, then they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise).  Rewarded selections also present a running total of earnings on the bottom of the screen.  If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise).  The timing of the reversal is based on a performance criterion.  Of note, this criterion has been revised since the task was initially employed (the initial version of this task (DRLT) used a performance criterion of 9 out of 12 correctly-completed trials, computed as a moving window).  The current version of the task (DRLT_v2) uses a performance criterion of 5 consecutive correctly-completed trials.
 +
 +
=== Task Structure Detail ===
 +
''' This is what we had worked on before, but could use updating. We'd like to capture a schema that can handle each of the tasks in the CNP, so please think general when editing -fws'''
 +
 +
* Task Structure ''(please given an overview of the task procedures here [i.e., overall design, block, trial, and within-trial event structure and timing])''
 +
** The BART has a series of instructional screens at the outset, followed by the experimental session comprised of 40 balloon trials.
 +
*** Six instructional screens.  Screens are advanced with a right mouse click from the examiner. 
 +
**** 1.  Basic task description.
 +
**** 2. Placement of hands and keys used for pumping and stopping (mapped based on handedness).
 +
**** 3. Pumping description and appearance of balloon.
 +
**** 4. Explosion description and appearance of exploded balloon.
 +
**** 5. Reiteration of key used to stop pumping/cash in (mapped based on handedness).
 +
**** 6. Questions screen prior to beginning task.
 +
*** Experimental trials (40 trials).  All experimental trials are preceded by a fixation cross screen.
 +
**** Fixation cross (500 ms) screen with blank white background.
 +
**** Balloon trial 1. Red or blue balloon presented (randomized).
 +
***** Within-trial structure
 +
****** Pumping increases balloon size with successively enlarged images (self-paced).
 +
****** Participant cashes out or the balloon explodes.
 +
****** Total points earned screen presented for 1500 ms.
 +
**** Balloon trial 2 begins after a fixation cross screen, and this is sequence is repeated for 40 trials.
 +
*** End. Thank you screen presented.  Grand Total Points also displayed.
 +
** Timing
 +
*** Instruction screens are static until advanced by examiner with a right mouse click.
 +
*** Fixation cross screens (500 ms).
 +
*** Balloon images are static until participant presses the pump key. Pressing the pump key instantaneously presents the following balloon image (either a slightly larger balloon or an exploded balloon).
 +
*** Images of an exploded balloon are presented for 1500 ms, with Total Points text underneath.
 +
*** When the participant cashes out, just the Total Points text is presented for 1500 ms.
 +
*** The ending thank you screen is static until advanced by the examiner with right mouse click.
 +
 +
* Stimulus Characteristics
 +
**sensory modality: Visual. Balloons are either red or blue.  Balloon images start with a size of 154 X 154 pixels and are increased in height and width by 2 pixels with each pump.  The fixation cross is black text in size 18 bold font in Comic Sans MS.
 +
**functional modality: visuoperceptual and linguistic (understanding of text).
 +
**presentation modality: computer display, no audio, directions are assisted by examiner.
 +
 +
* Response Characteristics
 +
**responses required: left or right key press for pumping or cashing in.  Button mapping is based on handedness.
 +
***effector modality: Manual button press.
 +
***functional modality: Manual button press.
 +
**response options ''(e.g.,  yes/no, go/no-go, forced choice, multiple choice [specify n of options], free response)'': yes/no, pump or cashout.
 +
**response collection ''(e.g., examiner notes, keyboard, keypad, mouse, voice key, button press)'': Button press and recording of responses in Eprime 2.0.
 +
 +
=== Task Schematic ===
 +
 +
Schematic of the Balloon Analogue Risk Task
 +
 +
[[File:blue_001.JPG]]  [[File:blue_002.JPG]]  [[File:blue_003.JPG]]  (cash out)  Total Points: XX
 +
 +
[[File:red_001.JPG]]  [[File:red_002.JPG]]  [[File:red_exp_003.JPG]]  (explosion) Total Points: XX
 +
 +
=== Task Parameters Table ===
 +
 +
[[File:TaskParamTable.png]]
 +
 +
=== Stimuli ===
 +
Stimuli consist of red or blue balloons which begin with a size of 154 X 154 pixels and are increased in height and width by 2 pixels with each pump.  When a balloon explodes, an image comparable in size and color is displayed as a burst balloon for 1500 ms.
 +
 +
A white screen with a black fixation cross (size 18 bold font in Comic Sans MS) is presented for 500 ms before every balloon trial.
 +
 +
=== Dependent Variables ===
 +
The primary dependent variable is the mean number of pumps on trials in which the balloon did not explode; these have been termed '''adjusted pumps'''.  Adjusted pumps are preferred to absolute pumps because explosions artificially restrict the range of pumping behavior (for evidence of the bias associated with absolute pumps see Pleskac et al. 2008).
 +
 +
Because the red and blue balloons differ considerably in their probability of explosion, it may also be useful to analyze the mean adjusted pumps on the red and blue balloons separately.  In addition, because participants will likely adapt to the task over time, adjusted pumps for each balloon color on each quartile of trials (1-10, 11-20, 21-30, 31-40) can be analyzed to determine how performance changes over time. Note here that mean adjusted pumps for each quartile likely vary in terms of the number of trials averaged (due to explosions and the fact that balloon color is randomized across the task).  Other variables of interest may include the number of adjusted pumps on red and blue balloons which immediately follow an explosion, and the number of explosions for each balloon color.  Lastly, because variability in pumping behavior has been shown to be important in animal models (Jentsch et al., 2010), the coefficient of variation of adjusted pumps can be examined across balloons, and in the red and blue balloons separately (coefficient of variation = standard deviation of adjusted pumps divided by the mean adjusted pumps).
 +
 +
 +
Table of all available variables.
 +
 +
[[File:SST_Variables_Table.png]]
 +
 +
 +
=== Cleaning Rules ===
 +
 +
In checking the data, it is first important to determine if the task was completed.  This can be done by checking the pumps per trial data (CountNumTrial1, CountNumTrial2, etc.) for the 40 trials to make sure that a value is present for each trial (either -9998, 0, or another number).  It is not recommended to use data on tasks which are incomplete.  Secondly, the pumps per trial data can also be checked to see if any participants “cashed out” on multiple trials without ever pumping (this will produce a 0 for the trial; explosions are given -9998).  While this may happen on occasion by accident, participants with multiple “cash outs” without pumping should be considered for exclusion.  Lastly, the distribution of explosions and mean pumps can be checked across participants, and participants with very extreme behavior (e.g., > 3 standard deviations above or below the mean) can be considered for exclusion.
 +
 +
=== Code/Algorithms ===
 +
 +
Two scoring procedures are provided in the following text.  The first is a relatively inefficient method which derives variables from eprime files (specifically, E-Merge files), and then imports this data into SPSS to run two scripts for calculation of the final variables.  Using this method requires that you have eprime 2.0 and SPSS installed on your computer.  The second method is a more streamlined approach which derives the final variables from the individual eprime text files.  This latter method is used by the CNP biostatistics core.
 +
 +
E-prime to SPSS scoring method:
 +
 +
1) Open up e-prime studio, go to Tools, and click E-Merge.  Find the individual BART files you want to merge together in the left pane of the screen (displayed in the main screen), hold down Control and highlight them all.  Then hit the Merge tab. Save the new merge.  For some reason when you go to open that new merge file it may be empty, so go to open file and select it again to open it up.
 +
 +
2) In the merge file go to Tools and click Batch Analysis.  You can then select more than one of the pre-made scripts for scoring.  At the Batch window, hit Add Files, then find the Overall analyses by row script, and hit open.  Then find the Number of trials cash out with no pumps and the Mean Median SD RT data and add them.  Once they’re all there, hit the Excel tab and it will send all the results to an excel spreadsheet, with each one of the scripts represented on a different tab.  For some reason the Overall analyses by row page is called Presses by color each subject in excel, even though all the required data is there.
 +
 +
3) Delete the information at the top of the excel sheet so that you can import the data into SPSS.  When you go to open the excel sheet in SPSS, you can select which excel tab you want displayed.  You probably just want to open the Presses by color each subject (overall analyses) tab, which will provide you with the primary dependent variables (other variables, such as reaction time and the number of times a participant cashes out without pumping, are already available in the appropriate excel tab).
 +
 +
4) Once the overall analyses are in SPSS, open up the syntax called BART create additional variables for calculations.  This will make the additional variables you need to get the primary variables later.  After you run this syntax, make sure to DEFINE -9998 AS MISSING FOR EACH VARIABLE IN THE ENTIRE SPSS DATASET.  IF YOU DON’T DO THIS, ALL THE SUBSEQUENT CALUCATIONS WILL BE TOTALLY WRONG.
 +
 +
5) After you’ve run the above syntax, open up the BART calculate primary variables only after missing variables have been defined syntax, and run it.  This will create pretty much all the variables you need.
 +
 +
Here are the eprime scripts mentioned above:
 +
 +
 +
 +
------
 +
 +
History of Checking Scoring:
 +
 +
* Andy Dean independently checked Stone’s results in 2009 and 2010 and found his results to be accurate.
 +
* In April 2011, Andy Dean and Eugene Oh checked Stone’s results a second time.  All results were accurate, except for two issues.  1) The BART_RatioMeanRedtoBlueAdjPumps variable was being scored incorrectly by Stone’s script.  He fixed this error and it is now identical to that determined independently. 2) The BART_TotalPointsSession variable (the total points earned over the task) was found to be inconsistent with adjusted pumps.  While this variable should be the total adjusted pumps multiplied by 5, in many cases it was not.  This was determined to be an error in the eprime program itself, not an analysis error.  It appears that in some cases participants are still awarded points even though the balloon explodes.  We are looking into the specific cause of this issue.  In any case, it is imperative that the total points awarded not be used as a dependent variable.  Instead, adjusted pumps should be used, and this variable is accurate.
 +
 +
=== Data Distributions ===
 +
 +
 +
 +
 +
 +
 +
 +
 +
=== References ===
 +
 +
Dean, AC, Sugar, C, Hellemann, G, & London, E (2011) Is all risk bad? Young adult cigarette smokers fail to take adaptive risk in a laboratory decision-making test. Psychopharmacology [Epub ahead of print]
 +
 +
Jentsch JD, Woods JA, Groman SM, Seu E (2010) Behavioral characteristics and neural mechanisms mediating performance in a rodent version of the Balloon Analog Risk Task. Neuropsychopharmacology 35: 1797-806
 +
 +
Lejuez CW, Aklin WM, Jones HA, Richards JB, Strong DR, Kahler CW, Read JP (2003) The Balloon Analogue Risk Task (BART) differentiates smokers and nonsmokers. Exp Clin Psychopharmacol 11: 26-33
 +
 +
Lejuez CW, Read JP, Kahler CW, Richards JB, Ramsey SE, Stuart GL, Strong DR, Brown RA (2002) Evaluation of a behavioral measure of risk taking: the Balloon Analogue Risk Task (BART). J Exp Psychol Appl 8: 75-84
 +
 +
Pleskac TJ, Wallsten TS, Wang P, Lejuez CW (2008) Development of an automatic response mode to improve the clinical utility of sequential risk-taking tasks. Exp Clin Psychopharmacol 16: 555-64
 +
 +
Wallsten TS, Pleskac TJ, Lejuez CW (2005) Modeling behavior in a clinically diagnostic sequential risk-taking task. Psychol Rev 112: 862-80
 +
 +
 +
 +
 +
 
'''Basic Task Description'''
 
'''Basic Task Description'''
  

Revision as of 12:29, 27 May 2011

go back to HTAC


Basic Task Description

The deterministic reversal learning task (DRLT) is a computerized measure of reversal learning, a form of behavioral/cognitive flexibility that has been widely investigated in both animal and human studies (Robbins et al., 2000). This particular task version was modeled after a paradigm originally used to examine genetic associations with reversal learning in BXD strains of mice (Laughlin et al., 2011). Participants are presented with two doors and instructed that they will receive monetary rewards based on which door they choose. The initial side of reinforcement (right or left) is randomly selected by the computer. If participants select the door associated with the reward, they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise). Rewarded selections also present a running total of earnings on the bottom of the screen. If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise). Once participants have consistently selected the reinforced door, the reward contingencies for the doors are reversed. After this reversal, participants need to consistently select the newly reinforced door in order to complete the task. The primary dependent variable is the number of errors that are made after reversal.

Task Procedure

This DRLT was programmed in E-Prime 2.0. Participant responses are made on computer keyboard, with three keys marked for use in this task: A (left), H (middle), and '(right). At the beginning of the task, instructions are presented on the screen in yellow, size 20 Arial font, with a black screen background. The instructions are as follows:

"In this game, you have an opportunity to win money by

looking behind doors. Do your best to find as much money as possible.

You will actually be given the money that you earn in this task.

You may only use the index finger of your dominant hand to respond,

and you must respond as quickly as you can.

You may NOT use two hands on this task.

On the keyboard, you'll see three marked keys: left, middle, and right.

To start each play of the game,

PRESS AND HOLD THE MIDDLE KEY DOWN,

and release the key only when you see doors appear to the left and right.

Then choose one of the doors to open by pressing either the left or right keys.

You will get money based on which door you choose.

Before we start the game, let's take a moment to practice the key presses..."

Participants are then shown another instruction slide that prepares them for the practice trials:

"In the following trials you will practice making quick key presses.

You'll hear a sound to let you know if your responses are fast enough.

Please tell your experimenter when you are ready to begin the practice trials."

Practice block trials provide a basic orientation to the trial format and response keys in order to shape prompt responding during the task. Practice trials begin with a box in the center of the screen with the word "Press" and the words "Press and hold the middle button" at the top of the screen. Once the participant presses the middle key, the word "Hold" appears in the center box and the words "Wait until you see the doors!" appears at the top of the screen. The "Hold" cue remains on the screen for either 2000 ms or 2500 ms, during which time participants must keep the middle key pressed down. If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops! Make sure to wait for the doors. Let's try again." Following the "Hold" delay, two doors appear on the left and right sides of the screen. The doors remain on the screen for up to 4000 ms or until the participant presses the left or right key. While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key. If a left or right keypress is made within 2000 ms, then participants are given feedback of "Good Job!" in the center box, along with a positive noise (fanfare). If participants take too long to respond (more than 2000 ms), they receive the following feedback in the center of the screen: "Oops! Please respond more quickly. Let's try again." Feedback slides are all presented for a total of 2000 ms. After three consecutive responses within the 2000 ms time limit, the practice trials end and the following instructions are presented on the screen:

"Now that you are familiar with pressing the keys, we can start the game.

In this game, you'll need to choose either the left or the right door to win money.

Please tell your experimenter when you are ready to begin."

Experimental block trials follow general parameters similar to the practice trials, and begin with a box in the center of the screen with the word "Press." If participants do not press the middle button within 5000 ms, they receive the following feedback in the center of the screen: "Please press the middle key to begin the next trial." After pressing the middle key, the word "Hold" appears in the center box and remains on the screen for either 1000 ms, 1500 ms, 2000 ms, or 2500 ms. These 4 durations for the "Hold" cue are randomized across all trials. If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops! Make sure to wait for the doors. Let's try again." Following the "Hold" delay, two doors appear on the left and right sides of the screen. The doors remain on the screen for up to 2000 ms or until the participant presses the left or right key. While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key.

For each experimental block of trials, the computer randomly selects either the left or right door to be initially associated with reward. Participants receive $0.50 each time they select the rewarded door. As soon as the participant presses the key corresponding to the side of the rewarded door, then they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise). Rewarded selections also present a running total of earnings on the bottom of the screen. If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise). The timing of the reversal is based on a performance criterion. Of note, this criterion has been revised since the task was initially employed (the initial version of this task (DRLT) used a performance criterion of 9 out of 12 correctly-completed trials, computed as a moving window). The current version of the task (DRLT_v2) uses a performance criterion of 5 consecutive correctly-completed trials.

Task Structure Detail

This is what we had worked on before, but could use updating. We'd like to capture a schema that can handle each of the tasks in the CNP, so please think general when editing -fws

  • Task Structure (please given an overview of the task procedures here [i.e., overall design, block, trial, and within-trial event structure and timing])
    • The BART has a series of instructional screens at the outset, followed by the experimental session comprised of 40 balloon trials.
      • Six instructional screens. Screens are advanced with a right mouse click from the examiner.
        • 1. Basic task description.
        • 2. Placement of hands and keys used for pumping and stopping (mapped based on handedness).
        • 3. Pumping description and appearance of balloon.
        • 4. Explosion description and appearance of exploded balloon.
        • 5. Reiteration of key used to stop pumping/cash in (mapped based on handedness).
        • 6. Questions screen prior to beginning task.
      • Experimental trials (40 trials). All experimental trials are preceded by a fixation cross screen.
        • Fixation cross (500 ms) screen with blank white background.
        • Balloon trial 1. Red or blue balloon presented (randomized).
          • Within-trial structure
            • Pumping increases balloon size with successively enlarged images (self-paced).
            • Participant cashes out or the balloon explodes.
            • Total points earned screen presented for 1500 ms.
        • Balloon trial 2 begins after a fixation cross screen, and this is sequence is repeated for 40 trials.
      • End. Thank you screen presented. Grand Total Points also displayed.
    • Timing
      • Instruction screens are static until advanced by examiner with a right mouse click.
      • Fixation cross screens (500 ms).
      • Balloon images are static until participant presses the pump key. Pressing the pump key instantaneously presents the following balloon image (either a slightly larger balloon or an exploded balloon).
      • Images of an exploded balloon are presented for 1500 ms, with Total Points text underneath.
      • When the participant cashes out, just the Total Points text is presented for 1500 ms.
      • The ending thank you screen is static until advanced by the examiner with right mouse click.
  • Stimulus Characteristics
    • sensory modality: Visual. Balloons are either red or blue. Balloon images start with a size of 154 X 154 pixels and are increased in height and width by 2 pixels with each pump. The fixation cross is black text in size 18 bold font in Comic Sans MS.
    • functional modality: visuoperceptual and linguistic (understanding of text).
    • presentation modality: computer display, no audio, directions are assisted by examiner.
  • Response Characteristics
    • responses required: left or right key press for pumping or cashing in. Button mapping is based on handedness.
      • effector modality: Manual button press.
      • functional modality: Manual button press.
    • response options (e.g., yes/no, go/no-go, forced choice, multiple choice [specify n of options], free response): yes/no, pump or cashout.
    • response collection (e.g., examiner notes, keyboard, keypad, mouse, voice key, button press): Button press and recording of responses in Eprime 2.0.

Task Schematic

Schematic of the Balloon Analogue Risk Task

Blue 001.JPG Blue 002.JPG Blue 003.JPG (cash out) Total Points: XX

Red 001.JPG Red 002.JPG Red exp 003.JPG (explosion) Total Points: XX

Task Parameters Table

TaskParamTable.png

Stimuli

Stimuli consist of red or blue balloons which begin with a size of 154 X 154 pixels and are increased in height and width by 2 pixels with each pump. When a balloon explodes, an image comparable in size and color is displayed as a burst balloon for 1500 ms.

A white screen with a black fixation cross (size 18 bold font in Comic Sans MS) is presented for 500 ms before every balloon trial.

Dependent Variables

The primary dependent variable is the mean number of pumps on trials in which the balloon did not explode; these have been termed adjusted pumps. Adjusted pumps are preferred to absolute pumps because explosions artificially restrict the range of pumping behavior (for evidence of the bias associated with absolute pumps see Pleskac et al. 2008).

Because the red and blue balloons differ considerably in their probability of explosion, it may also be useful to analyze the mean adjusted pumps on the red and blue balloons separately. In addition, because participants will likely adapt to the task over time, adjusted pumps for each balloon color on each quartile of trials (1-10, 11-20, 21-30, 31-40) can be analyzed to determine how performance changes over time. Note here that mean adjusted pumps for each quartile likely vary in terms of the number of trials averaged (due to explosions and the fact that balloon color is randomized across the task). Other variables of interest may include the number of adjusted pumps on red and blue balloons which immediately follow an explosion, and the number of explosions for each balloon color. Lastly, because variability in pumping behavior has been shown to be important in animal models (Jentsch et al., 2010), the coefficient of variation of adjusted pumps can be examined across balloons, and in the red and blue balloons separately (coefficient of variation = standard deviation of adjusted pumps divided by the mean adjusted pumps).


Table of all available variables.

SST Variables Table.png


Cleaning Rules

In checking the data, it is first important to determine if the task was completed. This can be done by checking the pumps per trial data (CountNumTrial1, CountNumTrial2, etc.) for the 40 trials to make sure that a value is present for each trial (either -9998, 0, or another number). It is not recommended to use data on tasks which are incomplete. Secondly, the pumps per trial data can also be checked to see if any participants “cashed out” on multiple trials without ever pumping (this will produce a 0 for the trial; explosions are given -9998). While this may happen on occasion by accident, participants with multiple “cash outs” without pumping should be considered for exclusion. Lastly, the distribution of explosions and mean pumps can be checked across participants, and participants with very extreme behavior (e.g., > 3 standard deviations above or below the mean) can be considered for exclusion.

Code/Algorithms

Two scoring procedures are provided in the following text. The first is a relatively inefficient method which derives variables from eprime files (specifically, E-Merge files), and then imports this data into SPSS to run two scripts for calculation of the final variables. Using this method requires that you have eprime 2.0 and SPSS installed on your computer. The second method is a more streamlined approach which derives the final variables from the individual eprime text files. This latter method is used by the CNP biostatistics core.

E-prime to SPSS scoring method:

1) Open up e-prime studio, go to Tools, and click E-Merge. Find the individual BART files you want to merge together in the left pane of the screen (displayed in the main screen), hold down Control and highlight them all. Then hit the Merge tab. Save the new merge. For some reason when you go to open that new merge file it may be empty, so go to open file and select it again to open it up.

2) In the merge file go to Tools and click Batch Analysis. You can then select more than one of the pre-made scripts for scoring. At the Batch window, hit Add Files, then find the Overall analyses by row script, and hit open. Then find the Number of trials cash out with no pumps and the Mean Median SD RT data and add them. Once they’re all there, hit the Excel tab and it will send all the results to an excel spreadsheet, with each one of the scripts represented on a different tab. For some reason the Overall analyses by row page is called Presses by color each subject in excel, even though all the required data is there.

3) Delete the information at the top of the excel sheet so that you can import the data into SPSS. When you go to open the excel sheet in SPSS, you can select which excel tab you want displayed. You probably just want to open the Presses by color each subject (overall analyses) tab, which will provide you with the primary dependent variables (other variables, such as reaction time and the number of times a participant cashes out without pumping, are already available in the appropriate excel tab).

4) Once the overall analyses are in SPSS, open up the syntax called BART create additional variables for calculations. This will make the additional variables you need to get the primary variables later. After you run this syntax, make sure to DEFINE -9998 AS MISSING FOR EACH VARIABLE IN THE ENTIRE SPSS DATASET. IF YOU DON’T DO THIS, ALL THE SUBSEQUENT CALUCATIONS WILL BE TOTALLY WRONG.

5) After you’ve run the above syntax, open up the BART calculate primary variables only after missing variables have been defined syntax, and run it. This will create pretty much all the variables you need.

Here are the eprime scripts mentioned above:



History of Checking Scoring:

  • Andy Dean independently checked Stone’s results in 2009 and 2010 and found his results to be accurate.
  • In April 2011, Andy Dean and Eugene Oh checked Stone’s results a second time. All results were accurate, except for two issues. 1) The BART_RatioMeanRedtoBlueAdjPumps variable was being scored incorrectly by Stone’s script. He fixed this error and it is now identical to that determined independently. 2) The BART_TotalPointsSession variable (the total points earned over the task) was found to be inconsistent with adjusted pumps. While this variable should be the total adjusted pumps multiplied by 5, in many cases it was not. This was determined to be an error in the eprime program itself, not an analysis error. It appears that in some cases participants are still awarded points even though the balloon explodes. We are looking into the specific cause of this issue. In any case, it is imperative that the total points awarded not be used as a dependent variable. Instead, adjusted pumps should be used, and this variable is accurate.

Data Distributions

References

Dean, AC, Sugar, C, Hellemann, G, & London, E (2011) Is all risk bad? Young adult cigarette smokers fail to take adaptive risk in a laboratory decision-making test. Psychopharmacology [Epub ahead of print]

Jentsch JD, Woods JA, Groman SM, Seu E (2010) Behavioral characteristics and neural mechanisms mediating performance in a rodent version of the Balloon Analog Risk Task. Neuropsychopharmacology 35: 1797-806

Lejuez CW, Aklin WM, Jones HA, Richards JB, Strong DR, Kahler CW, Read JP (2003) The Balloon Analogue Risk Task (BART) differentiates smokers and nonsmokers. Exp Clin Psychopharmacol 11: 26-33

Lejuez CW, Read JP, Kahler CW, Richards JB, Ramsey SE, Stuart GL, Strong DR, Brown RA (2002) Evaluation of a behavioral measure of risk taking: the Balloon Analogue Risk Task (BART). J Exp Psychol Appl 8: 75-84

Pleskac TJ, Wallsten TS, Wang P, Lejuez CW (2008) Development of an automatic response mode to improve the clinical utility of sequential risk-taking tasks. Exp Clin Psychopharmacol 16: 555-64

Wallsten TS, Pleskac TJ, Lejuez CW (2005) Modeling behavior in a clinically diagnostic sequential risk-taking task. Psychol Rev 112: 862-80



Basic Task Description

The deterministic reversal learning task (DRLT) is a computerized measure of reversal learning, a form of behavioral/cognitive flexibility that has been widely investigated in both animal and human studies (Robbins et al., 2000). This particular task version was modeled after a paradigm originally used to examine genetic associations with reversal learning in BXD strains of mice (Laughlin et al., 2011). Participants are presented with two doors and instructed that they will receive monetary rewards based on which door they choose. The initial side of reinforcement (right or left) is randomly selected by the computer. If participants select the door associated with the reward, they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise). Rewarded selections also present a running total of earnings on the bottom of the screen. If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise). Once participants have consistently selected the reinforced door, the reward contingencies for the doors are reversed. After this reversal, participants need to consistently select the newly reinforced door in order to complete the task. The primary dependent variable is the number of errors that are made after reversal.

Task Procedure

This DRLT was programmed in E-Prime 2.0. Participant responses are made on computer keyboard, with three keys marked for use in this task: A (left), H (middle), and '(right). At the beginning of the task, instructions are presented on the screen in yellow, size 20 Arial font, with a black screen background. The instructions are as follows:

"In this game, you have an opportunity to win money by

looking behind doors. Do your best to find as much money as possible.

You will actually be given the money that you earn in this task.

You may only use the index finger of your dominant hand to respond,

and you must respond as quickly as you can.

You may NOT use two hands on this task.

On the keyboard, you'll see three marked keys: left, middle, and right.

To start each play of the game,

PRESS AND HOLD THE MIDDLE KEY DOWN,

and release the key only when you see doors appear to the left and right.

Then choose one of the doors to open by pressing either the left or right keys.

You will get money based on which door you choose.

Before we start the game, let's take a moment to practice the key presses..."

Participants are then shown another instruction slide that prepares them for the practice trials:

"In the following trials you will practice making quick key presses.

You'll hear a sound to let you know if your responses are fast enough.

Please tell your experimenter when you are ready to begin the practice trials."

Practice block trials provide a basic orientation to the trial format and response keys in order to shape prompt responding during the task. Practice trials begin with a box in the center of the screen with the word "Press" and the words "Press and hold the middle button" at the top of the screen. Once the participant presses the middle key, the word "Hold" appears in the center box and the words "Wait until you see the doors!" appears at the top of the screen. The "Hold" cue remains on the screen for either 2000 ms or 2500 ms, during which time participants must keep the middle key pressed down. If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops! Make sure to wait for the doors. Let's try again." Following the "Hold" delay, two doors appear on the left and right sides of the screen. The doors remain on the screen for up to 4000 ms or until the participant presses the left or right key. While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key. If a left or right keypress is made within 2000 ms, then participants are given feedback of "Good Job!" in the center box, along with a positive noise (fanfare). If participants take too long to respond (more than 2000 ms), they receive the following feedback in the center of the screen: "Oops! Please respond more quickly. Let's try again." Feedback slides are all presented for a total of 2000 ms. After three consecutive responses within the 2000 ms time limit, the practice trials end and the following instructions are presented on the screen:

"Now that you are familiar with pressing the keys, we can start the game.

In this game, you'll need to choose either the left or the right door to win money.

Please tell your experimenter when you are ready to begin."

Experimental block trials follow general parameters similar to the practice trials, and begin with a box in the center of the screen with the word "Press." If participants do not press the middle button within 5000 ms, they receive the following feedback in the center of the screen: "Please press the middle key to begin the next trial." After pressing the middle key, the word "Hold" appears in the center box and remains on the screen for either 1000 ms, 1500 ms, 2000 ms, or 2500 ms. These 4 durations for the "Hold" cue are randomized across all trials. If they release the middle key prematurely during this time, they receive the following feedback and the trial restarts: "Oops! Make sure to wait for the doors. Let's try again." Following the "Hold" delay, two doors appear on the left and right sides of the screen. The doors remain on the screen for up to 2000 ms or until the participant presses the left or right key. While the doors are present, the "Hold" cue is then replaced by "Left or Right?" which prompts the participants to press either the left or right key.

For each experimental block of trials, the computer randomly selects either the left or right door to be initially associated with reward. Participants receive $0.50 each time they select the rewarded door. As soon as the participant presses the key corresponding to the side of the rewarded door, then they immediately see two quarters appear in place of the door and hear a rewarding sound (cash register noise). Rewarded selections also present a running total of earnings on the bottom of the screen. If participants select the door that is not associated with the reward, they immediate see an empty white box in place of the door and hear a negative sound (buzzing noise). The timing of the reversal is based on a performance criterion. Of note, this criterion has been revised since the task was initially employed (the initial version of this task (DRLT) used a performance criterion of 9 out of 12 correctly-completed trials, computed as a moving window). The current version of the task (DRLT_v2) uses a performance criterion of 5 consecutive correctly-completed trials.

Task Structure Detail