jensroes / nonadjacent-sequence-learning

Learning nonadjacent sequences
Apache License 2.0
0 stars 0 forks source link

Paper - Method #10

Open Mark-Torrance opened 1 year ago

Mark-Torrance commented 1 year ago

I'm just finishing writing this. Could someone give me answers to the following:

jensroes commented 1 year ago

What is the physical height of the display that we used?

We used the wall mount monitor. I don't know the height. Is this something Sofia reported in her thesis? Otherwise I can measure it on Tuesday.

What instructions, as exactly as possible, did we give to participants in the three experiments?

See here. I think this is generally what our RAs did.

Did we recalibrate between blocks (i.e. between the adjacent and non-adjacent conditions)?

We recalibrated between blocks. However "between blocks" is the same as between adjacent and non-adjacent dependencies. In exp 1 and 2, adjacent and non-adjacent dependencies were presented in different blocks but in exp 3 both dependencies were presented at the same time, in both blocks.

Mark-Torrance commented 1 year ago

Visual angle calculations (just for reference)

#eye distance from monitor
a <- 500

# dot height as proportion of monitor height 
prpn_dot <- 1/20

# aoi height as proportion of monitor height
# aoi radius is 2.75 radius of dot
prpn_aoi <- 2.75 * prpn_dot

# monitor height (mm)   
h_display <- 380 # need to check this

# dot height
h_dot <- h_display * prpn_dot

# aoi height    
h_aoi <- h_display * prpn_aoi

#visual angle
# dist = distance of eye from monitor
# height = height of stimulus on screen
# returns visual angle subtended by stimulus
vangle <- function(dist, height){
  2*(atan((height/2)/dist)*180/pi)
}

# dot visual angle
vangle(a, h_dot)

# aoi visual angle
vangle(a, h_aoi)
Mark-Torrance commented 1 year ago

Here are draft methods sections. We need to add in monitor height, and if it's not 380 mm then recalculate visual angle.

sequlearn-nonadjacent_methods.docx

karlnicholls commented 1 year ago

Here are draft methods sections. We need to add in monitor height, and if it's not 380 mm then recalculate visual angle.

sequlearn-nonadjacent_methods.docx

It was an array of 16 dots, not 12.

Instructions I gave were pretty much as they were in the document Jens linked above ("Please follow these instructions exactly"). Only change I made was saying that they would "see an array of 16 dots", not "you will see an array of dots". I also added that the task was split into two blocks and that they would have a rest in the middle, if they wanted it, instead of waiting for them to finish the first block and then offering them a break.

Mark-Torrance commented 1 year ago

It was an array of 16 dots, not 12.

Arse. I thought that might be the case. I'll have to redo the figure. @karlnicholls or @jensroes can you point me towards the location files that you used? I also don't have the trial definition files. I don't actually need these for the methods section, but we should have then saved somewhere safe (other than on the eye tracking computer).

jensroes commented 1 year ago

My fault for not having populated this repository yet. Very clumsy :)

Location files: locations_2.xlsx but only LS1 and LS2

Trial definition files:

Mark-Torrance commented 1 year ago

Revised method, now featuring 16 dots, and a rather larger figure, which I think is clearer.

sequlearn-nonadjacent_methods.docx

jensroes commented 1 year ago

Just checked the height of the monitor:

# monitor height (mm)   
h_display <- 300 # need to check this

is the answer

Mark-Torrance commented 1 year ago

Revised version with correct monitor height and visual angles.

sequlearn-nonadjacent_methods.docx

jensroes commented 1 year ago

@Lai-Sang see last comment.