Laurel Ellen Brehm

Biography

Research

There is more than one way to convey most ideas using language: almost any picture could be described with multiple labels (it's a couch/ sofa, vs I have the same one from Ikea, vs it's Mom's favorite place to sit), or multiple grammatical forms (the dog is being chased by the cat/ the cat chases the dog; the staff is/ are on strike). This means that speakers must juggle multiple possible representations while preparing to speak, and listeners must be able to expect more than one candidate utterance from a speaker. A lot of language use also occurs in the context of conversation, which means that one often needs to plan an utterance while simultaneously listening to an interlocutor. This means that in order to be resilient speakers and listeners, we must be able to simultaneously represent multiple ideas and multiple linguistic plans.

My research focuses on these questions in language production and comprehension. I examine how individuals integrate multiple candidate representations of meaning (words, ideas) and form (syntax, phonology) through time with support from cognitive mechanisms such as memory and attention.

One line of my research focuses more directly on the interplay between production and comprehension, interfacing with the 'Double Act' departmental research cluster.

  • What do we mentally represent while listening and speaking in a single-task or in a conversational (dual-task) context?
  • How do speech errors reflect mental representations of meaning and structure?
  • How do listeners interpret errors or dialectically-driven variable forms?

 

Another line of my research focuses more directly on the domain-general cognitive mechanisms behind language, interfacing with the 'Learning, Memory and Adaptation' departmental research cluster.

  • What memory mechanisms does language recruit?
  • How does attention to scenes or auditory cues impact speech planning?
  • What attention / memory / social cognition mechanisms are in play in dyadic language contexts?

 

For more details, this is my CV.

Teaching

In my teaching, I strive to balance content with practical skills: in order to properly use a skill (e.g., critical thinking, statistical analysis, experimental design), one needs to situate it in a context in which it is useful.  See below for some recent courses I have taught.

2019-- Critical Peer Review.  Co-Instructor (with Sonja Vernes).
Max Planck Institute / IMPRS

2019-- Data Visualization. Instructor.
Max Planck Institute / IMPRS

2018-- Mixed Effect Models. Co-Instructor (with Phillip Alday).
Max Planck Institute / IMPRS

2018-- Data Visualization. Instructor.
Max Planck Institute / IMPRS

2017-- Language Learning from Two to Three. Instructor.
Pennsylvania State University, Language and Linguistics Day
Outreach program for 10th and 11th grade students

 

Tools: Scripts, stats and R tutorials

Tutorials

Data visualization

A four-session workshop on data visualization in R based upon perceptual principles. Sessions 1 and 2 focus on properties of good visualizations, and sessions 3 and 4 focus on implementation. Presented at MPI / IMPRS, April 2018.

Data Viz Day 1 (.pptx) -- Lecture: A crash course in visual perception & empirically derived dos and don'ts

Data Viz Day 2 (.pptx) -- Lecture: How to choose a plot / Plotting as story-telling; quick intro to tidyverse

Data Viz Day 3 (.R) -- Tutorial: Plots with unsummarized data: lines, points, heatmaps

Data Vis Day 4 (.R) -- Tutorial: Plots with summarized data / filled areas: error bars, boxplot, density plot, beeswarms, and animated plots.
 

A basic overview of ANOVA and mixed-effects models in R.

Targeted toward psycholinguists, with an emphasis on repeated measures analyses. Analysis uses Baayen’s LexDec data set. Presented at NU CSD R & stats workshop, 03/16/16

Link to Code (.R)
 

Contrast coding and weights in R; Fun with logistic mixed effects regression

Logistic regressions targeted towards psycholinguists, demonstrating contrast coding /centering data, and how to deal with non-converging models. Analyses use Titanic disaster surviorship and Baayen’s LexDec data set. Presented at NU CSD R & stats workshop, 03/17/16.

Link to Code (.R)
 

R grab-bag: Reporting MEM models, PCA, and bootstrapping techniques for confidence intervals and data analysis

A discussion of standards for reporting MEM models; principal components analysis; bootstrapping for confidence intervals and (non-parametric) hypothesis testing. Analyses use Baayen’s LexDec data set & data on the author’s poor parking choices. Presented at NU CSD R & stats workshop, 03/18/16.

Thumbnail


Link to Code (.R)
 

Software & Graphing

Piecewise regression for determining category structure of predictors

See Brehm & Goldrick (2017) 'Distinguishing discrete and gradient category structure in language'

Thumbnail


Functions (.R) --- Demo (.R)
 

R/Pantone Color Chart

A color chart of base R and Pantone named colors, ordered by hue and saturation:value, with hex and RGB codes.

Code to generate your own plot (.R) --- List of Pantone colors with hex and RGB values (.txt)

Thumbnail


User-defined orthogonal code checker

An Excel document for checking whether user-defined contrast codes are orthogonal.

Link to Workbook (.xlsx)

 

Press

Why do kids make certain types of speech errors?

A column for Babel magazine's "Ask A Linguist" feature. In English.
https://cloud.3dissue.com/18743/41457/106040/issue24/index.html?page=14
 

What are the tools we use for research on language production?

A YouTube video explaining how to 'see sound' (with spectrograms, obviously!). In Dutch.
https://www.youtube.com/watch?v=NBbqG18J4Mk

 

 

Share this page