Research/Teaching

Thesis Project - Cognitive Neuroscience

I worked on my honors thesis with my academic advisor, Dr. Jason Arndt (Middlebury College).
My research was about the production effect. Specifically, I was interested in resolving theoretical debates on whether item-order information from study lists plays a role in word recall performance in the production effect paradigm.
I investigated that the idea of "task-switching costs," which is extensively studied in another realm of the memory research field, should theoretically explain the mixed results observed in the past literature. I suggested that the cognitive demands of switching between different tasks at study in the production effect paradigm lead to impaired memory performance, particularly in words studied silently in the mixed-task list condition.

We have run the second experiment to follow up on my first experiment. (A big thanks to Professor Arndt and the RAs at Middlebury!) Altogether, we found evidence for the "global switching cost" (cognitive demands required to maintain multiple tasks throughout the task) playing a role, but not for the "local switching cost" (cognitive demands required to switch between tasks at encoding).

I presented a poster on this project at the 2023 Psychonomic Society's Annual Meeting.

Psychonomics23 poster 8.pdf

Independent Project - Psycholinguistics

I have worked on a psycholinguistics study titled "The Impact of Second Language Background on The Sensitivity of Emotional Prosody Identification," under the supervision of Prof. Brandon Baird.
Brief background:

When you want to convey your emotion in speech, you do not just change the choices of words or sentence structures...you may also change your intonation, tone, or pitch. Or, you could say it more loudly/silently, more breathy, slowly/quickly, or add some pauses. These paralinguistic features are collectively called, "prosody." And when it serves to convey the speaker's emotion, we call it "emotional prosody."
Emotional prosody is found to be, at least somewhat, universal. That is, regardless of language, there are shared features of acoustics in each type of emotion, and we all seem to have the ability to recognize emotion just from the prosody -- so even in foreign languages!
At the same time, however, emotional prosody is cultural-specific. People in the same culture/using the same language of the stimuli can identify emotion more accurately than those not in the same culture or using the same language. This indicates that knowing/experiencing the language might be facilitating the sensitivity of vocal emotion identification from emotional prosody.
A few research in the past have found that bilinguals can better recognize the emotion of the speaker just from prosody, compared to monolinguals, in the language they know.

Given these, I sought to explore how having a bilingual background (defined as proficiency/exposure in two languages in my experiment) can impact the extent of the sensitivity in emotional prosody identification.

I presented posters on this study at ILLS15 (15th Annual Meeting of the Illinois Language And Linguistics Society) and YLSS 2023 (Young Language Science Scholar Speaker Series at Penn State). Currently, I am working on post hoc acoustic analysis.

ILLS15 poster final draft (Yuka Tatsumi).pdf

Collaboration Project - Psycholinguistics

I worked with Dr. Monica Barbir (IRCN BabyLab at the University of Tokyo) on the "Mask Study" project, which was to investigate the role of facial cues on infant novel word learning, when I was interning for the Language Development Lab at RIKEN Center for Brain Science.
I contributed to piloting and leading all of the experiments (eye-tracker experiments for 12-, 21-, and 28-month-old children). I also reviewed literature and designed data analysis for the poster presentation for the 12th BCCCD Conference based on these results.

This project was presented as a talk at BULCD 2022.

The preliminary results of this new study were presented as a poster at Many Paths to Language (MPaL) 2023.

Collaboration Project - Linguistics/Speech Science

I am working on the "Surnames with Friends" project with Dr. Eric Pelzl (previously Post-doc for Dr. Janet van Hell's Bilingual and Linguistic Diversity Lab at Penn State; Currently an Associate Professor in the Department of Chinese and Bilingual Studies at The Hong Kong Polytechnic University), Dr. Navin Viswanathan (Penn State Communication Sciences and Disorders), and Dr. Anne Olmstead (Penn State Communication Sciences and Disorders).
This project explores how Mandarin Chinese speakers resolve the ambiguity caused by tone 3 sandhi in interactions, using surname + title phrases as stimuli (e.g., Lu3 "Tone 3 surname Lu" + Zhu3Ren4 "Director" is homophonous with Lu2 "Tone 2 surname Lu"+ Zhu3Ren4 "Director"). I am currently working on behavioral and acoustic data analysis to see what strategies participants used in an effort to resolve ambiguities in interaction.

We are presenting the results of this project at:

Psychonomics 2023
LSA 2024

LSA2024

Teaching Experiences

Guest Lectures at Penn State University:

LING100 Foundations of Linguistics (Spring 2023)

Spectrogram reading lecture (1 session)
Language Acquisition lecture (focus on phonetics and word acquisition; 1 session)

LING497 Special Topics -- Phonetic Analysis (Spring 2023)

Praat Scripting lecture (2 days)

Praat workshop for the CLS community (Fall 2022):

Beginners level
Intermediate/Advanced level

**Materials available on my GitHub

Previous Linguistics Projects

(***All of the projects below are my undergrad coursework, so they are not necessarily well-thought-out. This is merely the presentation of linguistic topics I enjoy studying!)

❶Pragmatism of "Yeah" by native and non-native English speakers: Corpus Study

As a final project in my sociolinguistics class, I worked on the different use of the discourse marker "yeah" in English conversation, using the Michigan Corpus of Academic Spoken English (MiCASE). I focused on the position of the yeah (turn-initial, turn-medial, and word-final) and its function. Based on my literature review, I concluded that the function of yeah can be classified into seven categories: 1) Backchanneling, 2) Agreement/"yes", 3) turn-taking, 4) repair, 5) progress check, 6) closing, and 7) passive recipiency.

Although previous conversational analysis research have argued that turn-medial "yeah" and repair "yeah" are specific to non-native speakers, these uses were found both in NS and NNS speech. This may indicate that the specific uses of "yeah" associated with non-native speech are merely a bias that can lead to misconceptions and unnecessary linguistic profiling.

❷Use of Japanese SFP "ね" [ne]: error by L2 learners of Japanese

Japanese "ね" ([ne]) is one of the most frequently used SFP (sentence final particle) by native speakers to indicate various meanings and intentions of the speaker. Due to its wide variety of semantic nuances, the SFP Ne is often "misused" or "overused" by L2 learners of Japanese.

The first part of the research was to classify its use by native speakers from the literature. In the second part, I used NINJAL L2 learners corpus to analyze the error patterns of SFP Ne by non-native speakers.

After a careful literature review, I classified its use by native speakers into 8 types:

1) request confirmation

2) seeking/showing agreement

3) showing agreement to the entire speech act (used alone)

4) comments on information in the addressee’s territory

5) draw the addressee’s attention; get sympathy o ideas speakers want to explain about themselves

6) speech softener　

7) Discourse marker- solicit Aizuchi(back-channeling expression)

8) attention getter

As a result of L2 errors analysis, (6) softener use and (3) agreement use were most frequent, while (4) territory use and (8) attention-getter use were misused the most in %.

The possible causes of errors are: the nonexistence of SFP in their L1s, exposure to the use of SFP Ne from example dialogue in textbooks or/and pop culture, language proficiency level in Japanese, and exposure to native speakers of Japanese.

Although there were some limitations, this research allowed me to summarize the use of SFP Ne, which has not done in past ltierature, and take one step further to classify common error use tendencies made by non-native speakers.

❸ Praat acoustics analysis of "Unknown" language

I worked on analyzing the acoustics data of Finnish with Praat and SPSS for Phonetics and Phonology course (where the stats part was not required but I did it for fun)

Previous Internships (2018)

①Social Psychology Research Intern

For freshman summer, I visited Dr. Yi-Ching Angel Liu's business psychology lab at National Taiwan University (NTU). I was involved in the study of virtual and cross-cultural team dynamics and team management.

Also, I conducted a project on the relationship between functional diversity and the performance of multicultural groups. We created a survey from scratch, asked NTU summer class students to answer the survey, conducted ANOVA analyses, and found that the functional diversity factor does not necessarily lead to better group performance in the short term.

Along with that, I also worked as her research assistant to help her with a literature review for an upcoming project on the motivation of team members in large international/cross-cultural organizations.

②Neuroscience Research Intern

I did an internship for Dr. Keigo Kohara at Kansai Medical University Neuroscience Laboratory in Osaka, Japan.

I mainly worked on the microscopy imaging techniques of trisynaptic circuits in the hippocampus of single-cell gene knockout mice. I performed perfusion fixation, intracranial injection of viral vectors, and brain slice preparation of adult mice.

Page updated

Google Sites

Report abuse