The main function is read_edf()
that imports an EDF file
(or throws an error, if EDF API is not installed). By default it will
import all events and attempt to extract standard events: saccades,
blinks, fixations, logged variables, etc.
Events and individual event types are stored as tables inside the
eyelinkRecording
object with trial
column
identifying individual trials. Trial 0 correspond to events (e.g,
DISPLAY_COORDS
message, etc.) sent to the eye tracker
before the first trial.
To import samples, add import_samples = TRUE
and,
optionally, specify which sample attributes need to be imported, e.g.,
sample_attributes = c('time', 'gx', 'gy')
. All attributes
are imported if sample_attributes
is not specified. See
package and EDF API documentation for the full list of attribute
names.
# import samples with all attributes
gaze <- read_edf('eyelink-recording.edf', import_samples = TRUE)
# import samples with selected attributes
gaze <- read_edf('eyelink-recording.edf',
import_samples = TRUE,
sample_attributes = c('time', 'gx', 'gy'))
You can convert makes samples cyclopean. In case of a binocular recording, a single column would represent an average between to eyes. E.g., \(gx = (gxL + gxR) / 2\). In case of a monocular recording, a single column will contain information for a recorded eye. This is particularly useful, if you have multiple monocular recordings but some for the right eye and some for the left.
The package includes various utility functions that need to be called
separately. For example, to parse non-standard events: recorded areas of
interest (extract_AOIs
) and trigger events that help to
time events (extract_triggers
).
or, tidyverse-style
library(tidyverse)
gaze <-
read_edf('eyelink-recording.edf') %>%
extract_AOIs() %>%
extract_triggers()
Function adjust_message_time()
alters time stamps of
messages that are prefixed by a numeric offset that makes
synchronization easier for e E-Prime / Presentation. I.e., a message
"-50 TARGET_ONSET"
that was sent at 105600
becomes a message "TARGET_ONSET"
and its time stamp is
105550
.