Skip to main content



Ph.D. candidate (2014 - current)

Learning Sciences
School of Education and Social Policy at Northwestern University

Dissertation Title: Reimagining research methods in an age of complexity: studying people’s reasoning of complex systems through agent-based-construction (a-b-c) interviews

Advisor: Uri Wilensky, PhD

Dissertation Abstract

Thinking about complex systems plays a critical role in our lives such as personal decision making (e.g., relocating, investing), informed citizenry (e.g., civic dialogue, social justice, environmental protection), and career development (e.g., engineering, humanities, medicine). Therefore, investigating people’s natural ways of reasoning about real-world complex systems holds the key to developing an analytical framework that can inform further research on designing learning environments, technologies, and interventions. However, the very properties that make these systems difficult to think about also make it harder for researchers to expose people’s reasoning processes. Prior work on this topic is dense with studies on cognitive challenges and misconceptions, which often characterize novice and expert ways of reasoning about complex systems as a dichotomy. Such theories overlook the great diversity in people’s reasoning and target a wholesome replacement of novice learners’ mental models.

In my dissertation study, I propose a novel research method called agent-based-construction (a-b-c) interviews. In an a-b-c interview, the interviewer and the interviewee collaborate on developing a dynamic computational model; the interviewee provides the ideas, and the interviewer writes the corresponding code in the NetLogo agent-based modeling environment. In other words, the interviewer probes the participant’s thinking by writing simple pieces of code in addition to asking clarifying questions. The primary hypothesis of my dissertation is that a-b-c interviewing would expose participants’ reasoning about complex systems at a fine-grained and spontaneous level that may be inaccessible through research methods used in prior studies (e.g., clinical interviewing, concept mapping, questionnaires) because agent-based modeling offers representational affordances that could allow participants to express their thinking in a piece-by-piece manner. I present the results of two empirical studies on conducting a-b-c interviews with adult participants.

My findings indicate that this computationally mediated approach exposes participants’ reasoning about complex systems as dynamic conceptual ecologies with fine-grained knowledge elements that are spontaneously constructed and continuously restructured. By tracking the change in each participant’s thinking over the course of multiple a-b-c interviews, I find that novices’ mental models of complex systems include abundant idiosyncratic pieces that are not central in their final explanations but instrumental in the transition from primitive explanations to complex ones. I also present the results of a small-scale comparative study between a-b-c interviews and traditional clinical interviews based on verbal prompts.

My dissertation constitutes a first step in rethinking the prevalent novice-expert dichotomy in the Learning Sciences literature and building a framework for accommodating intermediate forms of knowledge and leveraging transition patterns in the design of future learning environments for teaching about complex systems.

M.Sc. (2010 - 2014)

Science and Mathematics Education
Bogazici University

Thesis Title: Fostering students’ learning of probability through video game programming

Advisor: Yavuz Akpinar, PhD

Thesis Abstract
I designed and developed a middle school curriculum based on video game programming activities; then, I investigated the effects of my curriculum on primary school students’ learning of probability concepts. First, I conducted a pilot study with 15 primary school students. Based on results of the pilot study, I revised my curriculum and conducted a main study was with 30 primary school students.

In both studies, the students learned and used Scratch blocks-based programming environment as a programming tool. In the first few weeks, they learned how to use Scratch with simple introductory projects. Then, I presented a probability-related question in the beginning of each lesson and each student developed their own unique Scratch project based on the question. The students’ projects can be found at <>

I collected qualitative and quantitative data using three different measurement tools: Probability Achievement Test (PAT), Reflective Thinking towards Problem Solving Scale (RTPSS) for collecting quantitative data, and Student Project Assessment Rubric (SPAR) for qualitative data. Analysis of the study data revealed that students were able to learn and use Scratch and develop probability related/based algorithms that generate ran- dom results successfully. I found a statistically significant improvement on students’ understanding of probability concepts before and after the intervention. However, the effect of the intervention on students’ reflective thinking towards problem solving was negligible and not statistically significant.

B.Sc. (2005 - 2010)

Computer Education & Educational Technology
Bogazici University


Research and Developer team member (2014 - current)

Center for Connected Learning and Computer-Based Modeling at Northwestern University

I have been a graduate member of our research lab since I joined Northwestern University as a PhD student. In addition to contributing to our lab as a researcher, I also was a member of our developer team from 2015 to 2020.

Graduate research assistant (2015 - 2021)

CT-STEM Research Project at Northwestern University

I designed and led research studies on making computational thinking practices (e.g., coding, data analysis, modeling and simulation) as a staple of everyday science classrooms. I designed novice friendly blocks-based programming environments for science education. I worked with in-service high school teachers to create CT embedded science units. Created computational models of scientific phenomena. I investigated the sociocultural implications of embedding computational thinking activities into high school science units. I supervised undergraduate research assistants.

Startup founder & Software developer (2013-2014)

Odyssey R&D

I established and ran a startup company within an incubator space at Bogazici University and received two grants from Turkish government to develop prototypes of a gesture-controlled medical image viewer kit using Microsoft Kinect Developer Kit and Leap Motion Developer Kit (now called ultraleap).

I developed both prototypes in C#. The prototypes allowed physicians to load medical image sets such as CT scans and MRI images in DICOM format and displayed them in full screen. Once the image sets were loaded, doctors could use hand gestures to navigate between images and also manipulate images (e.g., zoom, rotate).

The use case for these prototypes was to help physicians navigate between images during medical operations, which was a known issue that occasionally slowed down medical operations because reviewing images using a mouse (a non-sterile device) meant taking off gloves, using the computer, resterilizing hands, and putting back gloves. Due to this limitation, many surgeons had assistants who followed their directions, which was still a suboptimal solution.

I collaborated with an assistant professor surgeon at Cerrahpasa Medical School in istanbul, who tested the prototypes I built. I also received a successful implementation grade at the final evaluation of both government grants. However, I decided to transfer the ownership of Odyssey R&D and my prototypes when I was accepted to my PhD program at Northwestern University.

Software Developer (2011 - 2013)

Istanbul Topkapi University (formerly Plato College of Design and Media)

I was the sole in-house software developer during the establishment of a small college that initially offered offered two-year associate degrees in programs such as graphics design, interactive media, and journalism. I developed and managed the college’s website using a LAMP stack. I also developed online platforms for some of the sub-departments of the college, mostly using PHP, HTML/JS/CSS, and MySQL.

  • I developed the webpage of the school’s independent art gallery PlatoSanat including a simple backend with PHP for the staff to publish upcoming exhibitions and performances.
  • I developed the online version of the school’s student newspaper using Wordpress.
  • I developed an internet TV channel called PlatoWebTV using a LAMP stack for front end and Brightcove as the video streaming service.
  • I set up the schools LMS platform using Moodle to deliver distance learning programs.

Graduate research assistant (2010 - 2013)

Bogazici University Foreign Languages Education Department

I was a research assistant for two state-funded research projects in the Foreign Languages Education Department. I primarily took on the role of software developer and quantitative data analyst for two doctoral thesis studies and one master’s thesis studies. All of the studies were funded by the Turkish government.

I developed a mobile phone application for data collection at classrooms, I set up and calibrated an ASL Eye Tracking device, I analyzed data generated during the eye tracking studies, and I also set up and managed an LMS to be used by high school teachers for another.

  • For the mobile application, I used Mono for Android (now Xamarin) and C#.
  • For the eye tracking studies, I used the proprietary software provided by ASL Eye Tracking.
  • For the LMS, I set up a Moodle instance with some customizations made with PHP.

Software Developer (2010 - 2011)

Istanbul Aydin University

I was part of a small team of 5 in-house developers. I contributed to the development of the school’s web-based class registration and planning platform (AYSIS) that included schedule planning, grades, and student registration.

  • I contributed to the main platform code using PhP and MySQL.
  • I created an iPhone app for students to be able to view their grades using MonoTouch (now Xamarin) and C#.
  • I created a touch-based kiosk interface for students to check their grades on the campus (mobile phones were not as prevalent back then) using C# Windows Forms.
  • I created a small utility application with C# Windows Forms to print student ID cards that interfaced with a magnetic card reader to load student information from the school’s database and also interfaced with a card printer to print student photo and information on the card.

Data Analyst Intern (2008 - 2009)


I developed a data analysis application using C# that parsed csv files generated by a B2B platform and analyzed the sales data to identify grocery stores with low sales performance. The team at Danone needed such an app to analyze this data because the B2B software they used back then only gave them raw data but no analysis tools. My app imported large sets of sales data, cleaned it, minimized it to preserve memory space, and then calculated some rudimentary descriptive statistics such as mean, median, standard deviation, and percentiles. Then, it generated a visual map of the grocery stores on a 2D layout and marked the stores that were at the bottom of the bell curve in terms of sales (µ - 2 * σ). Once such low performers were determined, the company engaged with those grocery stores to ensure that Danone’s products were placed on shelves properly.

Teaching Experience

Teaching Assistant (2018 - 2021)

School of Education and Social Policy at Northwestern University

  • LRN_SCI 426: Design of Technological Tools for Thinking & Learning
    • Instructor: Uri Wilensky
  • HDSP 435: Advanced Qualitative Methods
    • Instructor: James Spillane
  • HDC 351: Mapping and Spatial Analysis for Social Issues
    • David Uttal

Instructor (2011 - 2013)

Istanbul Topkapi University (formerly Plato College of Design and Media)

  • Introduction to Mathematics for Artists and Designers
  • Web Design and Development

Middle School Teacher (2013)

The Koc School

  • Applied Mathematics



American Educational Research Association (AERA)

American Psychological Association (APA)

Association for Computing Machinery

International Society of the Learning Sciences (ISLS)

National Association for Research in Science Teaching (NARST)

Manuscript Reviewer

ACM Transactions on Computer-Human Interaction

ACM Interaction Design and Children (IDC) Conference

ACM CHI Conference on Human Factors in Computing Systems

Journal of Science Education and Technology

Journal of Educational Computing Research

Science Education





Technical Skills


Primary Toolkit
Python PHP HTML/CSS Javascript NetLogo LaTex

LAMP Node.js Flask Wordpress GitHub Hugo Cloudflare Pages


Data Analysis

Qualitative Data Analysis
Atlas.ti NVivo

Text Mining
NLTK spaCy

Scikit-Learn PyTorch

Statistical Analysis
SPSS Pandas/Numpy

Data Visualization
Matplotlib Seaborn Chart.js

Design and Authoring

Photoshop Lightroom

Descript Markdown