Return to the Garner Interference Task page
						
									GARNER INTERFERENCE TASK
SCRIPT INFO

Script Author: Katja Borchert, Ph.D. (katjab@millisecond.com) for Millisecond Software, LLC
Date: 03-05-2013
last updated: 02-06-2019 by K.Borchert (katjab@millisecond.com) for Millisecond Software LLC

Script Copyright © 02-06-2019 Millisecond Software


BACKGROUND INFO

											*Purpose*
This script implements a Garner Interference Paradigm that systematically varies the dimensions of
irrelevant information to study interference on the processing of relevant information.

The implemented task is based on:

Ridderinkhof, K. R., van der Molen, M. W., Band, P. H., & Bashore, T. R. (1997). Sources of interference 
from irrelevant information: A developmental study. Journal of Experimental Child Psychology, 65, 315–341.(Experiment1)

Ridderinkhof et al (1997) used this paradigm to study the development of attention in children.


											  *Task*
The two main dimensions used in this paradigm are tilt and color. Sometimes participants have to 
respond to the tilt of a line, in other cases they respond to the color, either of a surrounding box 
(in which case color and tilt are SEPARATED dimensions) or the color of the line itself 
(in which case color and tilt are INTEGRATED dimensions => more interference predicted). 
In addition, the change in color and tilt is systematically varied. The change can be:
a) correlated (either the dimensions would require the same response=congruent OR they would require opposite responses=incongruent)
b) orthogonal (= independent of each other => absence of a correlation)
c) neutral (= change happens only in one dimension but never in the other).
																					  
In each trial, participants view a box in the middle of the screen. Inside the box is a line that can be tilted to the left
or right.  The color of the box or the color of the line itself can change from red to blue.

Participants are instructed to pay attention to only one dimension (=relevant dimension)
a) the tilt of the line OR b) the color of either the box (separated condition) or the line itself (integrated condition) 
and ignore changes in the other (irrelevant) dimension as best as they can. 

The actual task they have to do is a simple categorization task: if the line tilts to the right, press a certain key. 
If the line tilts to the left, press another key. Same for the color tasks.



DATA FILE INFORMATION: 
The default data stored in the data files are:

(1) Raw data file: 'GarnerInterferenceTask_raw*.iqdat' (a separate file for each participant)

build:							Inquisit build
computer.platform:				the platform the script was run on
date, time, subject, group:		date and time script was run with the current subject/groupnumber 
blockcode, blocknum:			the name and number of the current block
trialcode, trialnum: 			the name and number of the currently recorded trial
									(Note: not all trials that are run might record data but they will still be reflected by trialnum; 
									by default data is collected unless /recorddata = false is set for a particular trial/block) 
										
/blockcount:					counts the testblocks run										
/practice:						1 = practice blocks; 0 = experimental blocks
/dimension:						target dimension: 1 = color; 2 = tilt
/dim_config: 					1 = integrated; 2 = separated
/variationtype:					1 = orthogonal; 2 = correlated; 3 = control
/target:						"color1" vs. "color2" vs. "tilt1" vs. "tilt2"
/responsekeycongruence:			0 = congruence doesn't apply; 1 = color and tilt response keys congruent; 2 = color and tilt response keys incongruent
response:						the participant's response (scancode of response button)
correct:						the correctness of the response (1= correct; 0 = error)
latency: 						the response latency (in ms); measured from onset of stimuli
/ITI:							stores the current ITI (in ms)


(2) Summary data file: 'GarnerInterferenceTask_summary*.iqdat' (a separate file for each participant)

script.startdate:				date script was run
script.starttime:				time script was started
script.subjectid:				subject id number
script.groupid:					group id number
script.elapsedtime:				time it took to run script (in ms)
computer.platform:				the platform the script was run on
/completed:						0 = script was not completed (prematurely aborted); 1 = script was completed (all conditions run)
									
/overall_PropCorrect:			overall proportion correct (across all trials)
/overall_RT:					mean latency (in ms) of correct responses (across all trials)


/propCorrect_separated:			proportion correct responses across all 'Separated' trials
/meanRT_separated:				mean latency (in ms) of correct responses across all 'Separated' trials
/propCorrect_integrated:		proportion correct responses across all 'Integrated' trials
/mean corrRT integrated:		mean latency (in ms) of correct responses across all 'Integrated' trials
									=> Garner Interference would predict that you would see less interference in 'Separated' conditions
									than in 'Integrated' conditions; thus:
									higher proportion correct/faster mean latency in Separated Condition supports the presence 
									of Garner Interference in Integrated conditions
																				
/propCorrect_control:			proportion correct responses across all 'Control' trials
/meanRT_control:				mean latency (in ms) of correct responses across all 'Control' trials										
/propCorrect_ortho:				proportion correct responses across all 'Ortho' trials
/meanRT_ortho:					mean latency (in ms) of correct responses across all 'Ortho' trials	
/propCorrect_correlatedC:		proportion correct responses across all Congruent 'Correlated' trials
/meanRT_correlatedC:			mean latency (in ms) of correct responses across all Congruent 'Correlated' trials									
/propCorrect_correlatedIC:		proportion correct responses across all INcongruent 'Correlated' trials
/meanRT_correlatedIC:			mean latency (in ms) of correct responses across all INcongruent 'Correlated' trials
									=> The presence of Garner interference would predict that
									a) the (congruent) correlated dimension boosts performance 
									b) the orthogonal dimension hurts performance compared 
									to the control condition.

Separated summary scores for all individual conditions:
/propCorrect_color_ortho_separated:	 	proportion correct responses across all separated, orthogonal color trials
/meanRT_color_ortho_separated:			mean latency (in ms) of correct responses across all separated, orthogonal color trials
etc.	
	
EXPERIMENTAL DESIGN
2 Dimensions (Color vs. Tilt), blocked-format x 
2 Dimensional Configurations (integrated vs. separated), blocked-format x
3 Variation Types (control, correlated, orthogonal), blocked-format

(A) Color vs. Tilt: color changes (red <-> blue) vs. changes in the tilt of the line (left <-> right)
if relevant dimension is color: the color of the unattended configuration (line or box) is yellow
if relevant dimension is tilt and no color changes should occur at all (control condition): box and line are purple
(red+blue -> purple)
	
(B) Integrated vs. Separated: color changes of the line itself vs. color changes of the box surrounding the line

(C) Control vs. correlated vs. orthogonal:
Control: only changes in the relevant dimension occur; the irrelevant dimension is held constant
Correlated: changes in the relevant dimension are tied to (dependent on) changes in the irrelevant dimension
	2 possible matches: 
		congruent  -> both changes require the same response or
		incongruent -> both changes require opposite responses
Orthogonal: changes in the relevant dimension are independent of changes in the irrelevant dimension
(half of the combinations are congruent vs. half are incongruent)

=> 16 different block types, tested all within-subjects

EXPERIMENTAL SET-UP
(a) 4 practice blocks (with error feedback)
(b) 12 experimental blocks: each block consisted of 48 trials; left/right responses occur equally often (no feedback)
Note: you can easily reduce the number of trials run per block (see Editable Parameters)
(c) A rest block in between each block

Notes:
1. The number of trials run per block can easily be reduced (see Editable Parameters)
2. To run a reduced number of blocks, change the blocks called by the 'expt' elements under section EXPERIMENT


RESPONSE KEYS/STIMULI:
Use of response keys (default: Y and B keys) is counterbalanced across participants. Assignment is done by groupnumber to 
any of the 4 possible key combination conditions.

EVEN groupnumbers align the tilt with the location of the left and right response buttons.
ODD groupnumbers use the opposite location.

!Note: The location of the (default) response keys (right "L" or left "A" on the keyboard) could introduce the potentially 
unwanted variable  "compatibility with tilt direction". 
To avoid, use 
(a) different response keys (change under Editable Parameters, e.g. Y and B)
(b) different stimuli: A set of possible alternative stimuli is provided under section Editable Stimuli -> item.lines
The set of alternative stimuli are Vs that differ in whether they are upside down or rightside up (replacing the direction
of left vs right). The composite stimulus for the color control condition is a diamond as it contains both V directions.


INSTRUCTIONS
can be edited under section Editable Instructions

TASK DURATION
the default set-up of the script takes appr. 1 hour to complete

Notes:
1. The number of trials run per block can easily be reduced (see Editable Parameters)
2. To run a reduced number of blocks, change the blocks called by the 'expt' elements under section EXPERIMENT


EDITABLE CODE:
check below for (relatively) easily editable parameters, stimuli, instructions etc. 
Keep in mind that you can use this script as a template and therefore always "mess" with the entire code to further customize your experiment.

The parameters you can change are:

/line_fontheight:				sets the height of the lines in % screen height (default: 20%)

/responsekey1:					sets the scancode of the response key 1 (here: 30)
/responsekey1_label:			the label of response key 1 (default: "A")
/responsekey2: 					sets the scancode of the response key 2 (here: 38)
/responsekey2_label:            the label of response key 2 (default: "L")

/color1:						sets the first target color (here: red)
/color2:						sets the second target color (here: blue)
/alternativecolor1:				sets the first alternative color (here: yellow)
/alternativecolor2:				sets the second alternative color (here: purple), used in control
								condition when color is held constant and only tilt is changing

/responsetimeout:				sets how much time participants have to respond in ms (here: 5000ms)

/regularresttime:				rest time after each block in ms (default: 60000s)
/prolongedresttime:				rest time during specified long rest blocks in ms (default: 300000)
/prolongedrest:					determines after how many blocks a prolonged rest block should be given (default:10)


/nrTrialsPerPracticeblock:		the number of trials per practice block that should be run (default: 48)
									Note: 48 is the maximum that will run
/nrTrialsPerTestblock:			the number of trials per test block that should be run (default: 48)
									Note: 48 is the maximum that will run

Copyright © Millisecond Software. All rights reserved.
Contact | Terms of Service | Privacy Statement | Security Statement | GDPR
Website Security Test