Construal in language: a visual-world approach to the effects of linguistic alternations on event perception and conception

This repository contains scripts for the analyses reported in

Divjak, D., Milin, P., and Medimorec, S. (2020). Construal in language: A visual-world approach to the effects of linguistic alternations on event perception and conception. Cognitive Linguistics, 31(1).

The paper-package is split into R script files which are hosted here on GitHub and the supporting datasets in csv format are available at University of Birmingham UBIRA eData platform: UBIRA eData link.

The study contains several analyses, as is typical for studies that report eye-movement data. Each part of the analysis is kept in a separate R script file and a corresponding data file. Script and data files parallel the analytic steps in the following order:

Paper section	R scripts	Data files
3.1 Pre-analysing the effect of the experimental manipulation	3.1_ExperimentalManipulation.R	preposition_path_data.csv voice_path_data.csv dative_path_data.csv preposition_pupil_data.csv voice_pupil_data.csv dative_pupil_data.csv
3.2 Order of access	3.2_OrderOfAccess.R	preposition_order_data.csv voice_order_data.csv dative_order_data.csv
3.3 First run gaze duration 3.4 Total gaze duration per interest area	3.3-4_Durations.R	preposition_first_dwell_data.csv voice_first_dwell_data.csv dative_first_dwell_data.csv preposition_total_dwell_data.csv voice_total_dwell_data.csv dative_total_dwell_data.csv
Supplemental materials B	SuplB_CombinedAnalysis.R	combined_analysis.csv

Data description

Dependent variables

path_power_transformed - power-transformed (Box-Cox) gaze path length

pupil_power_transformed - power-transformed (Box-Cox) average pupil size

order_of_access - order of access of the respective interest area

first_dwell_power_transformed - power-transformed (Box-Cox) first run dwell time (i.e., gaze duration)

total_dwell_power_transformed - power-transformed (Box-Cox) total dwell duration (i.e., total fixation duration)

Independent Variables

TrialScaled - scaled (z-transformed) order of the trial

Category - category of constructions (natural for naturalistic viewing; PREPOSITION: typical and atypical; VOICE: active and passive; DATIVE: NP and PP)

CanonicalFirst - whether the canonical construction (typical, active, NP) is presented before its atypical partner (atypical, passive, PP); this is a binary or dummy coded variable

InterestArea - interest area labels (A, B, C)

Combined independent variables

InterestAreaAndCategory - interaction (combination) of the interest area and the category

Random Effect Variables

Item - experimental items (image/sentence)

Participant - participant ID

Combined random effect variables

ParticipantAndInterestAreaPosition - interaction (combination) of participant and position of the interest area (e.g., U, D, L, R, NO)

---

For further inquiries contact: [email protected]