Weapons ID Task

Technical Manual

Script Author: Katja Borchert, Ph.D. (katjab@millisecond.com), Millisecond

Created: January 04, 2017

Last Modified: January 31, 2023 by K. Borchert (katjab@millisecond.com), Millisecond

Script Copyright © Millisecond Software, LLC

Background

This script implements the Weapon ID Task as described in:

Payne, B.K. (2001). Prejudice and Perception: The Role of Automatic and Controlled Processes in Misperceiving a Weapon. Journal of Personality and Social Psychology, Vol. 81, 181-192.

Https://bkpayne.web.unc.edu/research-materials/

Notes: (1) This script runs the Weapon ID Task without a response deadline. (2) This script is written in such a way, that primes and targets can easily be substituted (see section Editable Stimuli). (3) This script template makes the following prime-target compatibility-assumptions* (relevant if primes and targets are changed): compatible pairings: prime1-target1; prime2-target2 incompatible pairings: prime1-target2; prime2-target1

•Payne (2001): the original Weapons ID Task explicitly states that the 'Black faces - weapons' condition is compatible whereas the 'Black faces - tool' condition is incompatible but does not make any explicit pairing assumptions about the 'White faces' conditions.

Duration

7 minutes

Description

Categorization task of images as gun (target1) or tool (target2) after seeing either a black face (prime1) or a white face (prime2). The targets (gun or tool) are masked after 200ms (default) and participants have to press one key if the target was a gun ('E') and another if it was a tool (e.g 'I'). In this script, there is no response deadline for the response.

Procedure

Design:
2 prime types (here: male black faces vs. male white faces) X 2 target types (here: tool vs. gun), tested within participants

1. Practice: learning to categorize the targets correctly
- minimum of 4 blocks of 12 trials (=all 12 targets), order of targets is randomized
If proportion correct is < targetPracticePerformance (0.90, default) during the last practice block, additional practice blocks
automatically start until the criterium of targetPracticePerformance is achieved
(or until the max. number of practice blocks has run; default: 10)
- feedback is provided: if incorrect response -> the image of the target is presented again for further study
- participants only advance to the test if they reach targetPracticePerformance during the last practice block run

Payne (2001) presented 48 practice trials without a specified 'targetPracticePerformance' that had to be reached

Practice Trial Sequence:
target (200ms) -> mask (until response) -> feedback (reinforces learning by presenting the same target once more for error response) until spacebar press -> ITI (500ms)


2. Test
- 192 trials = each of the four 'prime type x target type' combinations runs 48 trials (order of trials is randomized)
• per 'prime type x target type' combination:
- each prime is presented 4 times, order is randomized
- each target is presented 8 times, order is randomized
- Pairing:
each randomly selected face prime is paired with a randomly selected target of the appropriate category
(no constraints applied on the random pairing of primes and targets)
Example: in trial.prime1_target2: each of 12 black face primes are randomly paired with a tool target

Test Trial Sequence:
prime (200ms) -> target (200ms) -> mask (until response) -> ITI (500ms)


Key Assignment:
key assignment to tool/gun is counterbalanced by groupnumber
odd groupnumbers: 'gun' with right key on keyboard ("E") vs. 'tool' with left key on keyboard ("I")
even groupnumbers: 'tool' with right key on keyboard ("E") vs. 'gun' with left key on keyboard ("I")

Estimates of Controlled/Automatic Components:
In this script the focus is on gun (target1) identifications following different primes
congruent condition: prime1-target1 pairs (black faces preceding guns) => allows the estimate of hit rate for gun identifications
incongruent condition: prime1-target2 pairs (black faces preceding tools) => allows the estimate of false alarm rate for gun identification

Stimuli

https://bkpayne.web.unc.edu/research-materials/
- can be edited under section Editable Stimuli

Instructions

provided by Millisecond - they are not original to Payne (2001)
The instructions can be easily edited under section Editable Instructions

Summary Data

File Name: weaponsidtask_summary*.iqdat

Data Fields

NameDescription
inquisit.version Inquisit version number
computer.platform Device platform: win | mac |ios | android
startDate Date the session was run
startTime Time the session was run
subjectId Participant ID
groupId Group number
sessionId Session number
elapsedTime Session duration in ms
completed 0 = Test was not completed
1 = Test was completed
primeDuration Duration (in ms) of prime presentation (default: 500ms = SOA)
! if primeDuration was set to be longer than SOA, the script automatically set the primeDuration = SOA
soa Stimulus onset asynchrony (in ms) of prime and target (default: 500ms)
list.practiceAcc.mean Stores the mean practice performance of the last/final practice block
countPracticeBlocks Stores the number of practice blocks run
advance 0 = no, script was aborted after practice due to low categorization performance during practice
1 = participant advanced to test session
the following summary variables are based on ALL responses regardless of accuracy and timing constraints
propCorrectAll Proportion correct of ALL trial responses
meanRTAll Mean latency of ALL trials
efficiency Proportion correct of all trial responses in relation to the average response time
= propCorrectAll/meanRTAll
! for the remaining summary variables
only responses with minRT <= latencies <= maxRT are considered for accuracy and latencies statistics
propCorrect Overall proportion correct
meanRT Overall mean latency (in ms) of correct responses
meanlnRT Mean of the ln-transformed latencies of correct responses across all conditions
accPrime1Target1 Proportion correct responses in prime1 -> target1 (here: black -> gun condition)
rtPrime1Target1 Mean latency (in ms) of correct responses in prime1 -> target1 (here: black -> gun condition)
lnRTPrime1Target1 Mean of the ln-transformed latencies of correct responses in prime1 -> target1 (here: black -> gun condition)
accPrime1Target2 Proportion correct responses in prime1 -> target2 (here: black -> tool condition)
rtPrime1Target2 Mean latency (in ms) of correct responses in prime1 -> target2 (here: black -> tool condition)
lnRTPrime1Target2 Mean of the ln-transformed latencies of correct responses in prime1 -> target2 (here: black -> tool condition)
accPrime2Target1 Proportion correct responses in prime2 -> target1 (here: white -> gun condition)
rtPrime2Target1 Mean latency (in ms) of correct responses in prime2 -> target1 (here: white -> gun condition)
lnRTPrime2Target1 Mean of the ln-transformed latencies of correct responses in prime2 -> target1 (here: white -> gun condition)
accPrime2Target2 Proportion correct responses in prime2 -> target2 (here: white -> tool condition)
rtPrime2Target2 Mean latency (in ms) of correct responses in prime2 -> target2 (here: white -> tool condition)
lnRTPrime2Target2 Mean of the ln-transformed latencies of correct responses in prime2 -> target2 (here: white -> tool condition)
"list.ACC_prime1target1.itemCount", "list.ACC_prime1target2.itemCount",
"list.ACC_prime2target1.itemCount", "list.ACC_prime2target2.itemCount": store the number of data points considered for summary statistics

Raw Data

File Name: weaponsidtask_raw*.iqdat

Data Fields

NameDescription
build Inquisit version number
computer.platform Device platform: win | mac |ios | android
date Date the session was run
time Time the session was run
subject Participant ID
group Group number
session Session number
blockCode Name of the current block
blockNum Number of the current block
trialCode Name of the current trial
trialNum Number of the current trial
primeDuration Duration (in ms) of prime presentation (default: 500ms = SOA)
! if primeDuration was set to be longer than SOA, the script automatically
set the primeDuration = SOA
soa Stimulus onset asynchrony (in ms) of prime and target (default: 500ms)
compatibility 1 = prime-target pairing compatible
0 = prime-target pairing incompatible
This script template makes the following prime-target compatibility-assumptions
compatible pairings: prime1-target1; prime2-target2
incompatible pairings: prime1-target2; prime2-target1
primeType Here: "black" vs. "white"
targetType Here: "GUN" vs. "TOOL"
prime Stores the currently presented prime image
target Stores the currently presented target image
response The participant's response: scancode of response button (default: 18 = E; 23 = I)
18 = E = left
23 = I = right
responseCategory Translation of scancode into response category (here: "tool", "gun")
correct Accuracy of response: 1 = correct response; 0 = otherwise
latency The response latency (in ms); measured from: onset of target until response
list.practiceAcc.mean Stores the mean practice performance of the current block
countPracticeBlocks Stores the number of practice blocks run

Parameters

The procedure can be adjusted by setting the following parameters.

NameDescriptionDefault
minPracticeBlocks The minimum number of practice blocks run 4
maxPracticeBlocks The maximum number of practice blocks run before the script aborts because categorization performance is too low 10
targetPracticePerformance Proportion correct that participant has to reach during last practice block to continue on to task
to eliminate a limit, set targetPracticePerformance = -1
90%
soa Stimulus onset asynchrony (in ms) of prime and target 200ms
primeDuration Duration (in ms) of prime presentation
! if primeDuration is set to be longer than SOA, the script automatically sets the primeDuration = SOA
200ms = SOA
targetDuration Duration (in ms) of target before it is replaced by a mask 200ms
iti Duration (in ms) of intertrial interval after response and before a new trial starts 500ms
readyDuration Duration (in ms) of the 'get ready' trial 5000ms
primeSize The size of the prime pictures proportional to canvas height 40%
targetSize The size of the target pictures proportional to canvas height
responseKeys
20%
leftKey The left response key (here: E)
rightKey The right response key (here: I)
the assignment of left and right response key to tool vs. gun is counterbalanced by groupnumber
odd groupnumbers: left response key is used to identify gun, right response key is used for tool
even groupnumbers: right response key is used to identify gun, left response key is used for tool
RT cut-offs for summary statistics
minRT Latencies < minRT are not considered for summary statistics 100ms
maxRT Latencies > maxRT are not considered for summary statistics 1000ms