Task 2

Task 2 - Images working memory

Outline

This task presents participants with an array of stimuli (different pictures), followed by a probe stimulus (a picture). Participants have to indicate whether the probe stimulus appeared in the array or not, and how certain they are of their answer.
This task has two optional components, namely feedback and meta-capactiy prompt.

1. As the name suggests the feedback component provides participants with feedback on their response. By default this feedback is only on half of the trials (the other half have the same inter-trial interval with a blank screen).
2. The meta-capacity prompt asks participants to estimate their working memory after they have seen the array but before they have seen the probe (this condition is called "number"). In addition, there are two control conditions: "letter" and "delay", see the description of the metaOptions parameter for mor infomartion about each.

Config

run

Flag to include/exclude task from overall experiment

practice

Note

Not currently used

Flag to include/exclude practice trials

nImages

Integer specifying total number of images available

nPracImages

Note

Not currently used

Integer specifying total number of images available for practice

nPositions

Integer indicating the number of different potential positions available for stimuli to appear in. Also determines the spacing angle between the stimuli (e.g. 8 positions = 45 degree spacing, 10 positions = 36 degree spacing, 12 positions = 30 degree spacing).
These positions are arranged in a circle around the centre of the screen. They are equally spaced and the distance from the centre is defined by the radius parameter (see below).

radius

Distance in pixels of the centre of each the stimuli from the centre of the screen.

radiusJitter

Array of 2 integers indicating the bounds of jitter for the radius. 1st value is the lower bound and second is the upper bound. e.g. if radius is set to 100 and radiusJitter is set to [-10, 10] then the acutal value of radius will be somewhere on a uniform distribution between 90 and 110.

angleJitter

Array of 2 integers indicating the bounds of jitter for the angle. 1st value is the lower bound and second is the upper bound. e.g. if the angle is set at 45 degrees and angleJitter is set to [-2, 2] then the actual angle will be somewhere on a uniform distribution between 43 and 47.

fixationDuration

Duration (in ms) the fixation screen is displayed for

fixationPostTrial

Duration (ms) of blank screen after the fixation screen

consolidationTime

Duration (ms) of blank screen after the target array screen

stimulusDims

Array contianing the dimensions of the stimuli in pixels: [w, h].

nStimuli

Number of items presented on the target array screen. Maximum 8. Can be a single number (e.g. 5) or an array of numbers of items (e.g. [2, 6])

timePerItem

Duration (in ms) of the target array screen. Can be a single number (e.g. 500) or an array of numbers of items (e.g. [250, 500])

novelProbe

Caution

Can be changed, but not advised.

Whether the probe presented was in the target array. An array of two values: [true, false].

nTrialReps

Number of times the list of trials will be repeated. The length of the list of trials (and therefore the overall number of trials) will vary depending on a number of factors:

• The length of the nStimuli array
• The length of the timePerItem array
• THe length of the consolidationTime array
• The length of the probePresent array
• If the metaCapacity flag is set to true, the length of the metaOptions array
• If the feedback flag is set to true the length of the showFeedbackOptions array

e.g.

1. If the length of the nStimuli, timePerItem, consolidationTime, and probePresent arrays are all $2$ and both the metaCapacity and feedback flags are set to false. Then the length of the list of trials will be: $2 * 2 * 2 * 2 = 16$ and so if nTrialReps is set to $5$ then the total number of trials will be: $5 * 16 = 80$.
2. If the length of the nStimuli and timePerItem arrays are both $1$, the probePresent and consolidationTime array lengths are $2$, the feedback flag is false, and the metaCapacity flag is true with a length of $3$ for the metaOptions array. Then the length of the trial list will be: $1 * 1 * 2 * 2 * 3 = 12$, and if the nTrialReps is set to $5$ then the total number of trials will be: $12 * 5 = 60$.

responseOptions

Array of labels for the different response options

metaCapacity

Flag setting whether meta-capacity prompts are presented in the trial

metaOptions

Array of different meta-capacity prompts to be used:

• "number": Prompt asking participants to indicate how many items they think they will be able to recall
• "letter": Prompt asking participants to find and click in the letter A
• "delay": Add a delay in place of the prompt

metaLetters

Array of letters that can be presented in the "letters" prompt.

metaDuration

Duration (ms) of the meta-capacity prompts.

feedback

Flag setting whether feedback is presented after the trial

showFeedbackOptions

Array of options for showing feedback after a trial.
By default set to: [false, true] meaning that feedback is shown after a trial 50% of the time. Adjusting the number of true and false values will adjust the frequency that feedback appears, but will also increase the number of trials unless other parameters are also adjusted (e.g. nTrialReps).

feedbackDuration

Duration (ms) of the feedback screen.

feedbackImgs

Array of images to use for the feedback screen. Contains two values: [image_for_positive_feedback, image_for_negative_feedback]

adaptiveEncode / adaptiveConsol

These sections of the configuration file control if and how the adaptive version of the task works.
One object (adaptiveEncode) controls the adaptive encoding section, and the other (adaptiveConsol) controls the adaptive consolidataion section. These can be active simultaneously.

Warning

The adaptive duration will be set by the score from Task 1, and will not run if Task 1 has not been run.

In the adaptive version of the task the trials wil either be a mixture of adaptive and default trials (i.e. 50% will use an adaptive duration, and 50% will use the default duration), or just adaptive trials (i.e. All trials will use the adaptive duration). By default is is the former 50:50 adaptive:default, but can be controlled using the adaptivePlaceholder parameter.

use

Flag setting whether the adaptive version of the task should be used

regTime

Default duration (ms) per item for adaptive encoding/consolidation.

adaptTimes

Adaptive durations (ms) per item for adaptive encoding/consolidation.
This is made up of a object containing key-value pairs. THey key is the score the participant received from Task1 and the value is the duration to use.
e.g. 2: 700 specifies that if the participant got a score of 2 then they would have a duration of 700 ms per item.

adaptLookBack

Number of trials to use when determining score from Task 1

adaptivePlaceholder

Array specifying type of trials to use in adaptive version of the task. Consists of either 1 or 2 values, either "adapt" and/or "reg".
"adapt" trials will use the adaptive duration (adaptTimes), while "reg" trials will use the default duration (regTime).
By default this will be ["adapt", "reg"] which will give 50% adaptive and 50% regular trials.

Important

If either of the use flags in adaptiveEncode or adaptiveConsol are set then the length of this array will be used in place of the timePerItem and consolidationTime arrays respectively.

Data

Data for task 2 are recorded in rows with Screen == probe and Task == 2
Columns of interest include:

• novel_probe: Was the probe presented novel (i.e. not in the memory array)? [true or false]
• probe_seen: Did the participant think the probe was in the array? [true or false]
• expected_response: The expected correct response [0: not seen or 1: seen]
• accuracy: Was the response given correct [0: no or 1: yes]
• nItems: The number of targets present on screen
• Encode_time: Value of encoding time in ms
• Consol_time: Value of consolidation time in ms
• RT: Reaction time of the response [in ms]
• response_confidence: Participant's confidence in their response [0: Sure, 1: Think, 2: Guess]