Short-term memory storage can be divided into separate subsystems for verbal information and visual information, and recent studies have begun to delineate the neural substrates of these working-memory systems. Although the verbal storage system has been well characterized, the storage capacity of visual working memory has not yet been established for simple, suprathreshold features or for conjunctions of features. Here we demonstrate that it is possible to retain information about only four colours or orientations in visual working memory at one time. However, it is also possible to retain both the colour and the orientation of four objects, indicating that visual working memory stores integrated objects rather than individual features. Indeed, objects defined by a conjunction of four features can be retained in working memory just as well as single-feature objects, allowing sixteen individual features to be retained when distributed across four objects. Thus, the capacity of visual working memory must be understood in terms of integrated objects rather than individual features, which places significant constraints on cognitive and neurobiological models of the temporary storage of visual information (Luck & Vogel, 1997, p.1).
Original paper
Paradigm is fully implemented into OpenSesame in Python 2.7 working and tested under Windows 7 SP2 64-bit and macOS 10.12.4.
In this paradigm we present some subset of complexity based on previous behavioral pilot study as 3, 4 and 6, 8 boxes.
First administrator is supposed to run NIC software to record EEG signal. It has to be run on the same machine since it is hard-coded as localhost at port 1234.
When connection at host:port is connected, done by socket Python module. In any case it would be in any Python distribution, otherwise install it by pip install socket in terminal.
Initial screen is dedicated for administrator with information if connection for EEG synchronization was established or not.
In both cases it requires administrators key press to continue despite success or error. In case of error administrator is suppose to:
- Shut down experiment
- Inform participant that he/she need to adjust conenction
- Close OpenSesame and NIC software
- Turn off and on the StarStim EEG system
- Re-open NIC software
- Check bluetooth synchronization information in NIC with EEG
- Check battery level
- Check if port at localhost is open (in right down corner)
- Re-open OpenSesame
- Start experiment again
- In any other cases (errors or crashes) feel free to contact me for support (ideally send me some image and caption what is wrong)
After synchronization procedure participant is asked to continue with any key press when he/she reads the main instruction. It is highly recommended to speak with the subject, to inform them what is going to happen and describe basics of their task and importance of clear signal.
Stimuli can occur on the screen with 10% margins from edges. It is pseudo-random order, e.g. once random possition for boxes etc. generated, then it is all the same for all subjects.
Experiment has 4 levels of complexity: 3, 4, 6 and 8 boxes (squares) which changes only 1 box color in 50% of all 120 trials in following ratio:
| Complexity | Number of trials |
|---|---|
| 3 boxes | 20 trials |
| 4 boxes | 20 trials |
| 6 boxes | 40 trials |
| 8 boxes | 40 trials |
Boxes coordinates are generated randomly only once (pseudo-random order), same as color and property if they change in color or not. Color are used copletely randomly same as order or changing.
It is used only these eight colors:
- orange
- grey
- purple
- red
- blue
- white
- green
- yellow
Size of the box is strictly set up as 64 x 64 px at 1920 x 1080 px resolution.
This is a sample logfile
It is collected only mouse pressed coded as LMB = 1 and RMB = 3 in the way of correctness in logfile which is automaticaly generated.
Each trial consists of intial cue image where some boxes appears (3, 4, 6 and 8) wich some pre-defined colors. After that there is retention screen which tends to minimize visual processing from intial cue and led participant to "forget" previous stimulation in term of short-term visual memory. Then it goes a second stimulation screen which presents a stimulation screen with the same layout and number of boxes but in 50% of all occurance it chanches one box color.
After stimulation participant is asked if it was or if it was not different - fisrt stimulation vs second stimulation. Then blank screen before next trial is presented.
| Screen | Description | Duration (ms) | EEG tag | Experiment dur |
|---|---|---|---|---|
| Box 1 | Intial cue | 100 | No | 12 000 ms |
| Retention | Retention screen - blank | 900 | No | 108 000 ms |
| Box 2 | Target stimulation - tagged | 1000 | Yes | 120 000 ms |
| Answer | Answering screen - response | Unlimited | Yes | Aprox 120 000 ms |
| Blank | Inter stimulus interval | 400 | No | 48 000 ms |
This is optimal event-related study timing due to study needs, e.g. as short as possbile. It takes in total 6,8 minutes (408 000 ms). You can see weight in total experiment duration in collumn Experiment dur in table above.
It is tagged only Box 2 screen and it is followring response (Answer screen). In offline analysis it will be used correctness and coded it into Box 2 screen. It means that is participant answers on 4 boxes and EEG got tag int 4, then when participant was correct it will be 14 (+10) or if wrong it will become 24 (+20).
For better undertanding it will be at the final stage of analysis transferend into human readable strings.
| Event | Tag integer |
|---|---|
| Box 2 screen with 3 boxes | 3 |
| Box 2 screen with 4 boxes | 4 |
| Box 2 screen with 5 boxes | 5 |
| Box 2 screen with 6 boxes | 6 |
| Correct answer | 10 |
| Incorrect answer | 20 |
In offline analysis these numbers become:
| Event | Tag integer |
|---|---|
| Box 2 screen with 3 boxes correct asnwer | 13 |
| Box 2 screen with 3 boxes incorrect asnwer | 23 |
| Box 2 screen with 4 boxes correct asnwer | 14 |
| Box 2 screen with 4 boxes incorrect asnwer | 24 |
| Box 2 screen with 6 boxes correct asnwer | 16 |
| Box 2 screen with 6 boxes incorrect asnwer | 26 |
| Box 2 screen with 8 boxes correct asnwer | 18 |
| Box 2 screen with 8 boxes incorrect asnwer | 28 |
EEG tag for StarStim is based on socket Python 2.7.12 module.
Full implementation of sending correct/incorrect tag is followings:
import socket
try:
starstim = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
starstim.connect(('localhost', 1234))
if var.correct == 1:
starstim.send("<TRIGGER>" + str(var.nBoxes + 10) + "</TRIGGER>")
else:
starstim.send("<TRIGGER>" + str(var.nBoxes + 20) + "</TRIGGER>")
except:
print 'Cannot connect to EEG system, please restart recording!'If is not conencted from some reason to EEG paradigm displays red font warining screen and shuts down the experiment before its start to prvent no tagged EEG recording.