In this project we investigated how Brazilian adults deal with multiple statistics to discover words in fluent speech. We created speech streams with transitional and phonotactic probabilities that work together, helping segmentation, or collide, hindering segmentation. Turns out that the combination of these statistics is really important for segmentation. The article is available on PsyArXiv and is published on Memory & Cognition.

In this project we investigated the famous "cognitive advantage" that bilingual infants are said to posses. We ran two studies of our own (with 7 and 20-months-old) and we also re-analyzed three open datasets. Overall, we found a tradeoff between speed of learning and flexibility to update learning. Our findings points to adaptations rather than advantages for both mono and bilinguals. The preprint is available at the PsyArXiv and the article is under review.

ManyBabies is a collaborative project for replication and best practices in developmental psychology research. It brings researchers together to address difficult outstanding theoretical and methodological questions about the nature of early development and how it is studied. I am involved in both leadership and research levels (MB-AtHome, MB4, MB5). Check here a cool shiny dashboard.

Pupillometry is awesome, but hard. When running pupillometry experiments it is necessary to control for the luminance of visual stimuli. In this project, I've adapted the SHINE MATLAB toolbox to control for several low-level properties of colorful images (SHINE_color)—which are much more attractive to infants than black and white stimuli. Obs. sample image from NOUN database.

In this project we are conducting a meta-analysis and systematic review of the empirical studies using the cross-situational word learning design. From 2007 up to 2019 we found 55 studies with hundreds of participants (from infants to older adults). Part of the meta-analysis, with infants/children, is openly available at Metalab (domain: "early language"; dataset: "cross-situational word learning"). We are now updating the review and running analysis on the adult data.

In this project we investigated whether explicit versus automatic reinforcement would have a greater effect on 4-years-old children's descriptions. I presented a picture to the participant and described it in passive voice, then I showed the participant a similar picture and asked she to describe it. If she described in passive voice, nothing happened, but if she described in active voice, explicit reinforcement, with favorite videos and compliments followed. What happened? Check the article at The Analysis of Verbal Behavior. :-)

In this workshop, I covered the process of pre-registering a study, including the reasons and means for doing so, as well as how to deal with deviations from a pre-registration. OSF Registries was used in a step-by-step demonstration. (Image: OSF)

In this lecture, I presented how open science practices can promote collaboration, diversity, and inclusion in Psychological Science. This lecture was part of a course on Open and Reproducible Research in Psychology offered by Krista Byers-Heinlein to undergraduate and graduate students at Concordia University, Montreal. (Image: Foster Open Science)

In this book chapter, we introduce recent research in statistical language learning. We cover four big topics: 1) Discovering the sounds in language, 2) Finding words in speech, 3) Mapping words to meaning, and 4) Learning rudimentary grammar. It offers a great and gentler introduction for undergraduate and graduate students.

In this lecture, I cover critical periods on language development, from the womb to the first anniversary. Some of the topics include how babies' perceptual attunement, speech segmentation, cross-situational word learning, and phonotactics. This lecture have been well received by undergraduate and graduate students from Brazil and Canada. (Image: Werker & Hensch, 2015)

In this semester-long course, I cover the principles of respondent (reflex) and operant learning. I present the variation & selection framework and discuss the concept of behavioral repertoire. I also cover some learning procedures such as stimulus pairing, shaping, reinforcement, punishment, extinction, and stimulus equivalence.

In this laboratory course, I introduce students to the selection by consequences framework in a practical way. We do a series of learning exercises with non-human subjects (lab rats) using the Skinner box. Some procedures that I cover are: simple and conditional discrimination, shaping, reinforcement, extinction, fading, and behavioral chaining, etc. (Image: Skinner, 1957)