Linguistic Serial Reaction Time Task - AGL-SRT:

Technical Manual

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

Created: January 14, 2023

Last Modified: January 11, 2025 by K. Borchert (katjab@millisecond.com), Millisecond

Background

This script implements Millisecond's version of the Linguistic Serial Reaction Time (SRT) task - also referred to as the AGL-SRT- described by Misyak et al (2010) and Quam et al (2018). The linguistic SRT task combines a serial reaction time task with artificial grammar learning (Misyak et al, 2010) and has been used as a measure of procedural memory in language learning (Quam et al, 2018).

References

Misyak, J. B., Christiansen, M. H., and Tomblin, J. B. (2010). On-line individual differences in statistical learning predict language processing. Front. Psychol. 1:31. doi: 10.3389/fpsyg.2010.00031

Quam, C., Wang, A., Maddox, W. T., Golisch, K., & Lotto, A. (2018). Procedural-Memory, Working-Memory, and Declarative-Memory Skills Are Each Associated With Dimensional Integration in Sound-Category Learning. Frontiers in Psychology, 9, 1828–1828. https://doi.org/10.3389/fpsyg.2018.01828

Duration

40 minutes

Description

Task

In this task, participants see a 2x3 matrix filled with nonwords.

Column1 holds two different 'A' nonwords (sampled from 3 possible ones). One 'A' word is the target and the other one is the foil.

Column2 holds two different 'B' nonwords (sampled from 24 possible ones). One 'B' word is the target and the other one is the foil.

Column3 holds two different 'C' nonwords (sampled from 3 possible ones). One 'C' word is the target and the other one is the foil.

For each column (starting with the A column), the computer plays a soundfile that pronounces the target and the participants' task is to select the spoken word in the matrix as fast as possible. As soon as the response is registered, the soundfile for the next column is played and so on. At the end of a sequence a blank screen is shown for a short duration before the next 2x3 matrix is presented.

Procedure

Participants work through 3 phases that flow into each other unannounced.

1. Training: Participants work through 6 blocks of training (72 trials each). During training, the C-nonwords always depend on the A-nonwords (the pairing of the three A-nonwords and the three C-nonword is randomly decided at script start) => artificial 'grammar learning'

2. Disruption: Participants work through 1 block of 24 trials. During the shorter disruption block, the previously learned AC-dependencies are completely avoided. Thus, A-nonwords are always presented with one of the C-nonwords that the A-nonwords were NOT paired with during training.

3. Recovery: Participants work through 1 block of 72 trials. During recovery, the training AC pairings are once again in effect.

Procedure

A-nonwords: the nonwords that appear in the left column of the matrix
B-nonwords: the nonwords that appear in the center column of the matrix
C-nonwords: the nonwords that appear in the right column of the matrix
(see list.A, list.B, list.C)

Preparation

At the beginning of the script run, the script randomly pairs the 3 A-nonwords with the 3 (non-adjacent) dependent
C nonwords. This (non-adjacent) dependency is then kept fixed for the entire training run.
Example: 'A = pel is paired with C = vok' throughout the training

The script also generates the 72 unique nonword strings:
=> 3 different, fixed non-adjacent dependency pairs (AC) x 24 different B nonwords = 72 unique nonword strings
Example: 'pel-wadim-vok'

Training Of 'Grammatical' Dependencies

By default this scripts runs 6 training blocks (see editable parameters to change number of training blocks.
Note that increasing the numbers requires adding additional summary variables)
- in this script each training block runs the 72 unique nonword strings consisting of the 3 nonwords (A,B,C)
once as a target and once as a foil
- the script ensures that each target/foil strings pairs run different nonwords in the A, B, and C positions.

1. this script generates a new list of target/foil pairs for every block. If no suitable pair list is generated
within given time constraints (unlikely), the script uses the default pairing lists instead
and leaves a note in the raw data file. The default target/foil pairings conform to all constraints.
The default target/foil pairs are fixed but the order of pairings is randomly determined, so even
if the default lists were used consecutively (highly unlikely), participants would not see the same order of trials.

2. this script uses the non-adjacent AC pairs for targets AND foils. The result is that each target sequence
is also run as a foil.

- the position of the A target (A foil) is randomly determined with the constraint that the A target
appears on top in 50% of all trials
- the position of the B target (B foil) is randomly determined with the constraint that the B target
appears on top in 50% of all trials
- the position of the C target (C foil) is randomly determined with the constraint that the C target
appears on top in 50% of all trials

positions for A,B,C targets are selected independently from each other

Disruption Of 'Grammatical' Dependencies

This scripts runs 1 disruption block (24 trials) of the grammatical dependencies AC.
The trained grammatical dependencies are disrupted for all pairs, thus A targets (A foils)
are never paired with their former dependant Cs.

- the disruption block runs 24 unique nonword strings consisting of the 3 nonwords (A,B,C)
once as a target and once as a foil
- the script ensures that each target/foil strings pairs run different nonwords in the A, B, and C positions.

1. If no suitable target/foil pair list is generated within given time constraints (unlikely),
the script uses a default pairing lists instead and leaves a note in the raw data file (see values.defaultSeq).
The default target/foil pairings conform to all constraints.
The default target/foil pairs are fixed but the order of pairings is randomly determined.

- the position of the A target (A foil) is randomly determined with the constraint that the A target
appears on top in 50% of all trials
- the position of the B target (B foil) is randomly determined with the constraint that the B target
appears on top in 50% of all trials
- the position of the C target (C foil) is randomly determined with the constraint that the C target
appears on top in 50% of all trials

positions for A,B,C targets are selected independently from each other

Stimuli

provided by Millisecond - can be edited under section 'Editable Stimuli/Lists'
The soundfiles were generated by FIVERR by (female American English Native Speaker), Millisecond .

Instructions

provided by Millisecond - can be edited under section 'Editable Instructions'

Summary Data

File Name: linguisticsrt_summary*.iqdat

Data Fields

Name	Description
inquisit.version	Inquisit version number
computer.platform	Device platform: win \| mac \|ios \| android
computer.touch	0 = device has no touchscreen capabilities; 1 = device has touchscreen capabilities
computer.hasKeyboard	0 = no external keyboard detected; 1 = external keyboard detected
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
sequenceFlag	Should be empty BUT sequenceFlag = stores the phase and blocknumber during which the default sequence was run because the script couldn't generate a unique sequence within time constraints
Training1-Training6
training1PropTargetA	Proportion selected A targets
training1MeanRTTargetA	Mean response time in selecting A targets
training1PropTargetB	Proportion selected B targets
training1MeanRTTargetB	Mean response time in selecting B targets
training1PropTargetC	Proportion selected C targets
training1MeanRTTargetC	Mean response time in selecting C targets
Disruption
disruptionPropTargetA	Proportion selected A targets
disruptionMeanRTTargetA	Mean response time in selecting A targets
disruptionPropTargetB	Proportion selected B targets
disruptionMeanRTTargetB	Mean response time in selecting B targets
disruptionPropTargetC	Proportion selected C targets
disruptionMeanRTTargetC	Mean response time in selecting C targets
Recovery
recoveryPropTargetA	Proportion selected A targets
recoveryMeanRTTargetA	Mean response time in selecting A targets
recoveryPropTargetB	Proportion selected B targets
recoveryMeanRTTargetB	Mean response time in selecting B targets
recoveryPropTargetC	Proportion selected C targets
recoveryMeanRTTargetC	Mean response time in selecting C targets

Raw Data

File Name: linguisticsrt_raw*.iqdat

Data Fields

Name	Description
build	Inquisit version number
computer.platform	Device platform: win \| mac \|ios \| android
computer.touch	0 = device has no touchscreen capabilities; 1 = device has touchscreen capabilities
computer.hasKeyboard	0 = no external keyboard detected; 1 = external keyboard detected
date	Date the session was run
time	Time the session was run
subject	Participant ID
group	Group number
session	Session number
blockcode	The name the current block (built-in Inquisit variable)
blocknum	The number of the current block (built-in Inquisit variable)
trialcode	The name of the currently recorded trial (built-in Inquisit variable)
trialnum	The number of the currently recorded trial (built-in Inquisit variable) trialnum is a built-in Inquisit variable; it counts all trials run even those that do not store data to the data file.
pairAC1	The AC1 pair (e.g. pel-tood)
pairAC2	The AC2 pair (e.g. dak-rud)
pairAC3	The AC3 pair (e.g. vot-jic)
a1	The A1 word (e.g. pel)
c1	The C1 word (e.g. tood)
a2	The A1 word (e.g. dak)
c2	The C1 word (e.g. rud)
a3	The A1 word (e.g. vot)
c3	The C1 word (e.g. jic)
phase	The current phase of the script (e.g. "training")
defaultSeq	0 = a unique (training) sequence could be generated for the current block 1 = a unique (training) sequence could not be generated; the default one was used instead (should be rare)
blockCounterPerPhase	The number of blocks run in the current phase
trialCounterPerBlock	The number of trials run in the current block
target	Stores the current target sequence (e.g. dak-loga-rud)
foil	Stores the current foil sequence (e.g. vot-kicey-jic)
aTarget	The A target (e.g. dak)
aTargetIndex	The itemnumber of targetA (used to call the soundfile)
aTargetSoundFile	The played targetA soundfile
bTarget	The B target (e.g. loga)
bTargetIndex	The itemnumber of targetB (used to call the soundfile)
bTargetSoundFile	The played targetB soundfile
cTarget	The C target (e.g. rud)
cTargetIndex	The itemnumber of targetB (used to call the soundfile)
cTargetSoundFile	The played targetC soundfile
aFoil	The A foil (e.g. vot)
bFoil	The B foil (e.g. kicey)
cFoil	The C foil (e.g. jic)
aTargetPosition	1 = target A on top; 2 = target A on bottom
bTargetPosition	1 = target B on top; 2 = target B on bottom
cTargetPosition	1 = target B on top; 2 = target B on bottom
response	The response of participant: the selected response (e.g. ATarget)
correct	Correctness of response (1 = correct, 0 = error)
latency	Response latency (in ms); measured from: onset of nonword

Parameters

The procedure can be adjusted by setting the following parameters.

Name	Description	Default
Design Elements
withBreaksAfterTrainingBlocks	False: no breaks after training blocks true: rest after training blocks	false
numberTrainingsBlocks	The number of training blocks run this script only collects summary data for up to 6 training block	6
Sizing Parameters
boxHeightPct	The proportional height of the grid boxes (relative to canvas height)	30%
Timing Parameters
startDurationMS	Delay (in ms) of the first nonword	250
itiMS	Intertrial interval (in ms) after each trial	750

Linguistic Serial Reaction Time Task - AGL-SRT:

Technical Manual

Background

References

Duration

Description

Task

Procedure

Procedure

Preparation

Training Of 'Grammatical' Dependencies

Disruption Of 'Grammatical' Dependencies

Stimuli

Instructions

Summary Data

Data Fields

Training1-Training6

Disruption

Recovery

Raw Data

Data Fields

Parameters

Design Elements

Sizing Parameters

Timing Parameters