Neurofeedback (NFB) task

Design

• 4 runs
  • Run1 = 11.2 minutes (671 seconds)
  • Run2 = 11.1 minutes (666 seconds)
  • Run3 = 11.2 minutes (673 seconds)
  • Run4 = 11.2 minutes (672 seconds)
  • TotalTime = 44.6 minutes
  • 32 Blocks per run
    • 4 conditions
      • Protocol 196KJ (A; 7 signals, 1 baseline)
      • Protocol 564D (B, 2 signals, 6 baseline)
      • Calibration I (C; 7 signals, 1 baseline)
      • Calibration II (D; 2 signals, 6 baseline)
    • 8 trials for each condition
    • 7 unique feedback signals for each run
      • B and D show signals that are subset of A and C signals
    • 6 unique feedback baselines for each run
      • A and C show baselines that are subset of B and D baselines
    • Each condition has a different fill color
      • Colors are counterbalanced across 4 different versions
    • 1 Block:
      • 4 seconds infusion:
        • 4 stages: 0, 33, 66, 100, 100
        • Each stage displays for 1 second
      • 2 seconds will improve
        • Keyboard responses are restricted to 1, 2, 3, 4, 5, 6, 7, 8, 9, 0
          • 1,2,3,4,5 = left value
          • 6,7,8,9,0 = right value
      • Random jitter
        • Jitter duration is randomly selected from an exponential distribution bounded between 0.33 - 2 seconds
        • 1 jitter length is sampled from the uniform distribution with bounds 4-6 seconds; this is to increase randomness
      • 10 seconds feedback
        • 3 stages
          • 0.25-2 seconds baseline (excluding the initial baseline screen)
          • 1.5-3.33 seconds ramp
          • rest at max
        • Feedback consists of Gaussian noise
        • For stages 2,3 two sine waves are added to signal
      • 2 seconds improved
        • Keyboard responses restricted to 1, 2, 3, 4, 5, 6, 7, 8, 9, 0
          • 1,2,3,4,5 = left value
          • 6,7,8,9,0 = right value
      • Random jitter
        • Jitter duration is randomly selected from an exponential distribution bounded between 0.33 - 2 seconds
        • 1 jitter length is sampled from the uniform distribution with bounds 4-6 seconds; this is to increase randomness
  • There are 10 seconds of baseline at the beginning and end of the paradigm; baseline displays only the background color
  • There is now a confirmation screen requiring input to continue to the next run. Pressing any character key will advance this screen.

Design Options

• Scan
  • 1 = Yes, this will make the presentations opaque
  • 2 = No, this will make the presentations translucent which is useful for debugging. If an error occurs in this mode, enter "sca" without quotes in the Matlab terminal to close the experiement.
• Participant ID: self explanatory; a directory Responses/[ID] will be created to store participant output
• StartRun: run to start
• EndRun: run to end
• Odd runs (1,3) display rates as YES/NO, even runs (2,4) display rates as NO/YES
• Testing
  • 1 = Yes, this will use NfbTestOrder.csv as the design which is signficiantly shorter for testing purposes
  • 2 = No, this will use NfbDesign.csv as the design which is what you want to use in the scanner
• Version controls the color scheme
• Color set controls what colors are displayed
  • Color set 1
    • Version 1
      • Protocol 196KJ = red
      • Protocol 564D = light blue
      • Calibration I = green
      • Calibration II = yellow
    • Version 2
      • Protocol 196KJ = light blue
      • Protocol 564D = red
      • Calibration I = yellow
      • Calibration II = green
    • Version 3
      • Protocol 196Kj = green
      • Protocol 564D = yellow
      • Calibration I = red
      • Calibration II = light blue
    • Verison 4
      • Protocol 196Kj = yellow
      • Protocol 564D = green
      • Calibration I = light blue
      • Calibration II = red
  • Color set 2
    • Version 1
      • Protocol 196KJ = orange
      • Protocol 564D = pink
      • Calibration I = purple
      • Calibration II = dark blue
    • Version 2
      • Protocol 196KJ = pink
      • Protocol 564D = orange
      • Calibration I = dark blue
      • Calibration II = purple
    • Version 3
      • Protocol 196KJ = purple
      • Protocol 564D = dark blue
      • Calibration I = orange
      • Calibration II = pink
    • Version 4
      • Protocol 196KJ = dark blue
      • Protocol 564D = purple
      • Calibration I = pink
      • Calibration II = orange

File/Directory Descriptions

• NfbTask.m - matlab script to run the social cognition task
• NfbDebug/Development/NfbDesign.csv - csv file listing the trials used in the experiment. Manually edit this file if you want to use a specific trial order, but make sure you keep the same format; otherwise, the task will not run.
• Fonts - directory containing the necessary fonts (not needed any more)
• NfbDebug/Development/CreateDesign.m - matlab script to create NfbDesign.csv file
• NfbDebug/Development/CreateWaveforms.m - matlab script to create feedback signals
• NfbDebug/Development/Waveforms.mat - mat file saving the feedback signals
• NfbResponses - participant response files are saved in this directory

Social Cognition (SC) Task

Design

• 2 runs selected from 5 different versions of the run
• Run1 = 9.8 minutes (590 seconds)
• Run2 = 9.9 minutes (598 seconds)
• Run3 = 9.9 minutes (592 seconds)
• Run4 = 9.9 minutes (596 seconds)
• Run5 = 9.9 minutes (592 seconds)
• Each version was generated using this script: http://www.bobspunt.com/easy-optimize-x/ which maximizes efficiency.
• 72 trials per run
  • 6 Conditions
    • 1: Pleasant - Happy (Congruent)
    • 2: Pleasant - Fearful (Incongruent)
    • 3: Pleasant - Neutral (Neutral)
    • 4: Unpleasant - Happy (Incongruent)
    • 5: Unpleasant - Fearful (Congruent)
    • 6: Unpleasant - Neutral (Neutral)
  • 12 trials for each condition in 1 run
  • appropriate images were randomly selected for each trial with sample
    • for example all Happy images including all genders were pooled together and then randomly assigned to conditions 1 and 4
  • 1 trial:
    • 3 seconds contextual picture
      • 72 unique pictures per run (144 total)
        • 36 pleasant per run
        • 36 unpleasant per run
    • 2 seconds face and rating
      • Keyboard responses are restricted to 1, 2, 3, 4, 5, 6, 7, 8, 9, 0
        • 1,2,3,4,5 = left response
        • 6,7,8,9,0 = right response
      • 36 male face per run (72 total)
        • 24 individuals
          • happy, fearful, neutral
          • 3 * 24 = 72
      • 36 female faces per run (72 total)
        • 18 individuals
          • happy, fearful, neutral
          • 3 * 18 = 54
          • some images are repeated
    • 2 - 6 seconds jittered inter-stimulus interval (ISI) sampled from an exponential distribution
      • Average of 3 seconds
      • min = 2 seconds
      • max = 5.59 seconds

Design Options

• Scan
  • 1 = Yes, this will make the presentations opaque
  • 2 = No, this will make the presentations translucent which is usefull for debugging. If an error occurs in this mode, enter "sca" without quotes in the Matlab terminal to close the experiemehnt.
• Participant ID: self explanatory; a directory Responses/[ID] will be created to store participant output
• Run1: selects what version to display for the first run (1-5)
• Run1 Order: determines the subset of images displayed. Typically this value should always be set at 1. There are 2 subset of images, one for each run. Order also controls the rating display screen locations. 1 = positive/negative, 2 = negative/positive
• Run2: selects what version to display for the second run (1-5). If this field is left blank, then only Run1 will be displayed.
• Run2 Order: determines the subsect of images displayed. Typically this value should always bet set at 2.
• Testing
  • 1 = Yes, this will use ScTestOrder.csv as the design which is signficiantly shorter for testing purposes
  • 2 = No, this will use ScDesign.csv as the design which is what you want to use in the scanner

File/Directory Descriptions

• ScTask.m - matlab script to run the social cognition task
• ScDebug/Development/FaceList.csv - csv file listing the images used in each order (run)
• ScDebug/Development/ScDesign.R - creates ScDesign.csv
• GetDevice.m - matlab file to list connected devices to computer. Use this to identify the DeviceIndex value for SocialCognitionTask. This is useless for Windows machines (confirm this statement).
• ScImages/Pleasant - directory containing pleasant contextual images
• ScImages/Unpleasant - directory containing unpleasant contextual images
• ScImages/Backgrounds - directory containing task background images
• ScImages/Faces - directory containing all face images
• ScResponses - participant response files are saved in this directory

Queries

• Queries/QueryDevices.m - matlab file to list connected devices to computer. Use this to identify the DeviceIndex value for SocialCognitionTask. This is useless for Windows machines (confirm this statement).
• Queries/QueryClose1.m - examines window close behavior with "sca"
• Queries/QueryClose2.m - examines window close behavior with "Screen('CloseAll');"
• Queries/QueryClose3.m - examines window close behavior with "Screen('Close');"
• Queries/QueryFeedback.m - examines feedback closely with manual movement
• Queries/QueryFont.m - queries if fonts properly display after being installed
• Queries/QueryKbInput.m - queries input from devices; displays the input on screen
• Queries/QueryLineLength.m - looks at line length relative to feedback background
• Queries/QueryScreen.m - queries display information
• Queries/QueryTiming.m - queries difference between actual and expected flip times
• Queries/QueryTrigger.m - queries trigger functionality
• Quereis/QueryWindowLength.m - queries window length by drawing a line

Scanner info

• screen reslution is 1024 x 768 (Will this be the in scanner display resolution?)
• operating system is Windows 7
• Response collection device: PST BRU 5 Buttons
• keys are probably mapped to home row not number pad, you can check matlab command prompt response outputs to confirm this