phd-interview/slides.tex
2026-04-23 15:09:16 +01:00

369 lines
14 KiB
TeX
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

\documentclass[aspectratio=169]{beamer}
\usepackage{tabularx}
\usepackage{booktabs}
\usepackage{csquotes}
\usepackage{graphicx}
\usepackage{xcolor}
\newcommand\todo[1]{\textcolor{red}{#1}}
% section frame
\AtBeginSection[]{
\begin{frame}
\vfill
\centering
\begin{beamercolorbox}[sep=8pt,center,shadow=true,rounded=true]{title}
\usebeamerfont{title}\insertsectionhead\par%
\end{beamercolorbox}
\vfill
\end{frame}
}
% URL coloring
\hypersetup {
colorlinks=true,
linkcolor=blue,
filecolor=magenta,
urlcolor=blue,
pdftitle={PhD Interview},
pdfpagemode=FullScreen,
}
% Referencing
\bibliographystyle{IEEEtran}
\setbeamertemplate{bibliography item}{\insertbiblabel}
\title{Can the use of a virtual avatar and scaffolding with AI agents improve
first-stage programming students' debugging ability?}
%\subtitle{aa}
\date{23rd of April 2026}
\author{Warwick New}
\begin{document}
\maketitle
%\section{What is the problem I'm trying to solve?}
\begin{frame}
\frametitle{What is the problem space?}
Novices tend to struggle with programming fundamentals, finding the basics
quite difficult to learn, and so introductory courses tend to have poor rates
of retention.
\newline
One factor influencing retention is students self-perception of their
programming ability. Negative perceptions can reduce retention and increase
the likelihood that students drop out in their first stage.
\cite{lewis2011}.
\newline
A particular point of frustration for novice programmers is attempting to
debug without thinking of the problem critically \cite[p.~23]{vickers2008}.
\newline
Previous interventions in this space have shown promise such as Gidget
\cite{Lee2014} that has used personification of compilers to great effect and
Metacodenition \cite{Pechorina2023} which helps the student create a mental
model about actions to take when a bug appears.
\newline
Recent literature has begun exploring the use of LLMs but out of the box,
found them in most cases to be worse than just reading the compiler.
\cite{Pechorina2023}.
%Recent literature has identified this as a problem space that AI
%can attempt to help resolve. Renzella et al's work \cite{renzella2025}
%demonstrates that there are aspects of learning that traditionally could only
%be fulfilled by an instructor outside highly controlled learning
%environments (like Gidget \cite{Lee2014}) that can be partially automated with
%generative AI. And therefore can be made more available to students outside
%timetabled sessions.
%%in this case reducing the cognitive load \cite{sweller1988} of reading and understanding compiler messages by providing more human-readable descriptions of compiler errors.
\end{frame}
%\begin{frame}
% \frametitle{What is the gap I've identified}
% \todo{Cite and Professionalise this slide}
% \begin{itemize}
% \item Much research into debugging education for students more recently is
% focused on generative AI models and some new paradigms that come from that
% \todo{\cite{}}. This is to be expected with a whole new paradigm in
% education in this area.
% \item Before this focus however much research into the area of first stage
% computer science debbugging was focused on teaching debugging in closed
% controlled environments \todo{\cite{}}, This allowed for experimentation
% with making fixing coding problems more accessible, in terms of compiler
% error messages and scaffolding of the processes a student might take when
% they ran into a problem.
% \item I think that we can take these classical educational tools and use
% them combined with more recent generative AI computing education
% technologies to.
% \begin{enumerate}
% \item Remove the need for locked down highly controlled debugging
% environments.
% \item Improve the human element to debugging messages and tools with how
% much more human the AI can appear to be to students.
% \end{enumerate}
% \end{itemize}
%\end{frame}
\begin{frame}
\frametitle{What is the Gap}
{\footnotesize
Two practices in helping students learn to debug have shown great promise when
it comes to improving self-efficacy and lowering programming anxiety. I
believe that we could use recent advances in AI to augment aspects of these
interventions.
\begin{itemize}
\item Personification of programming tools as fallible and encouraging, is
shown to have a positive impact on learning motivation and success as
shown in Gidget \cite{Lee2011}.
\item Scaffolding what steps to take when encountering a bug, improves
self-efficacy by lowering cognitive load as demonstrated Metacodenition
\cite{Pechorina2023}.
\end{itemize}
%Both of these works are restriceted to very specific environments limiting their application on real world coding projects.
\begin{figure}
\centering
\includegraphics[width=0.45\textwidth]{progprac}
\caption{Provided by Scott et al. \cite{Scott2014}}
\label{fig:question}
\end{figure}
This figure shows how programming self-concept influences programming anxiety
and increasing time on task, retaining more computing students.
}
\end{frame}
%\begin{frame}
% \frametitle{The fields this gap interacts with}
% \begin{itemize}
% \item Computer science education
% \begin{itemize}
% \item Specifically focusing on first stage debugging.
% \end{itemize}
% \item Psychology
% \begin{itemize}
% \item Student Self-efficacy. \todo{cite}
% \item Students engagement. (Time Spent working on programming tasks)
% \todo{Define engagement}
% \end{itemize}
% \item Artificial Intelligence
% \begin{itemize}
% \item Scaffolding of Generative AI queries to fit a more traditionally
% productive learning environment in a wide range of situations.
% \item Encourage personal reflection and action when attempting to solve
% an issue.
% \end{itemize}
% \end{itemize}
%\end{frame}
%\section{How I plan to address this space}
%\begin{frame}
% \frametitle{How am I addressing the gap}
% \todo{Talk about Gidget and offer scaffolding tools, then talk about how we
% can create a tool to aid intervention}
% I plan to create an intervention wherein I give students access to an AI tool
% that takes elements from previous interventions that align with the principles
% of improving self-efficacy in students, and combine it with more recent
% breakthroughs in LLM based technologies to allow the intervention to be used
% in a wider variety of programming environments and contexts.
%
% I then plan to evaluate its impact on self-efficacy and potentially other
% outcomes. \todo{Figure out what those outcomes are.}
%\end{frame}
%\begin{frame}
% \frametitle{The Intervention}
% \todo{Talk about methodology and the type of intervention that's taking place.
% Why this method over other methods etc}
%
% This artefact attempts to replicate the channels of self-efficacy improvement
% that one on one support from an instructor, by targeting these factors in
% Bandura's theory \todo{cite}:
%\begin{itemize}
% \item Mastery experiences (Experiencing more success in coding tasks).
% \item Verbal encouragement.
% \item Guided support.
%\end{itemize}
% \todo{Check the previous itemisation isn't mistaken}
%
% The agent will use techniques such as personification \todo{cite gidget} and
% an AI models ability to translate errors into more accessible descriptions
% replicate an instructors ability to help a student understand knowledge they
% may be missing in terms of the language of a compiler. The model will also
% focus on guiding the student towards the solution in a structured manner such
% as in \todo{cite structured debugging intervention paper}.
%
% The goal with this artefact is to as closely as possible replicate the
% experience and self-efficacy improvements that come with 1 on 1 tuition
% \todo{cite}, so that these benefits can be scaled up. And help seeking
% behaviours be made more available.
%\end{frame}
\begin{frame}
\frametitle{The Intervention}
I plan to create an artefact that uses scaffolding and personification in an
attempt to improve students self-efficacy \cite{Bandura1977} and reduce
cognitive load \cite{Sweller1988}.
\newline
This will be done by walking a student through any issues a real programming
environment can throw at them with the use of a conversational AI agent based
avatar, which will break down the process of solving debugging issues in their
own real world programming environment.
\newline
The key feature of the agent being that it has been created with pedagogic
strategy in mind, namely scaffolding and personality to see if AI can help
fill these gaps when structured effectively.
\end{frame}
\begin{frame}
\frametitle{Aims and objectives}
{\small
Question:
Can we use traditional pedagogic strategies to improve the
effectiveness of conversational AI agents in helping first stage
programming students believe in their ability to debug programs?
\newline
Aims:
\begin{itemize}
\item Improve first stage self perception on their ability to solve bugs.
\item Improve first stage students debugging ability. \newline
\end{itemize}
Objectives:
\begin{itemize}
\item Evaluate the changes in first stage programming students
self-efficacy.
\begin{itemize}
\item Both before and post intervention
\end{itemize}
\item Investigate factors that improve students self perceptions of their
programming ability.
\item To develop a tool that helps students improve at a rate that makes
their self perception of programming ability higher.
\item To develop a tool that students can fall back to when they can't seek
help elsewhere.
\end{itemize}
}
\end{frame}
\begin{frame}
\frametitle{Methodology}
Due to the nature of using a software tool to improve outcomes, an
intervention based experiment becomes necessary \cite[p.~242]{Coe2025}. The
projects' reliance on AI should also adhere to guidelines such as
CONSORT-AI \cite{Liu2020} in order to remain reproducible and transparent.
\newline
If ethical issues surrounding the withholding of resources from some students
that were made available to others even at random wasn't a worry, a random
selection of student participants would be the best approach to reduce bias
\cite[p.~245]{Coe2025}.
\newline
To counter this issue will I will be measuring results from a cohort before
the intervention has been fully developed and comparing results to a cohort
that has then since gained access to the artefact in a longitudinal study
\cite[p.~224]{Coe2025}.
\begin{itemize}
\item Year 1 -- Pre-intervention Cohort
\item Year 2 -- Intervention Cohort
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Data Collection}
%TODO come back to this
Self-efficacy is a psychological self-perception and as such must be collected
from communicating with the participants.
\newline
Several measurement instruments, typically questionnaires, have been developed
and validated for measuring self-beliefs in introductory programming contexts
(e.g., \cite{Scott2014}) and these are often complemented through thematic
analysis of post-intervention interviews \cite{Braun2006}.
%\newline
%
%Both of these methods require careful planning to avoid the influence of bias.
%I've interacted with them before in papers I've contributed to in the past
%\cite{Mitchell2021, Mitchell2022}, and I am confident in the GA's ability to
%guide me through the process of making sure I collect this data correctly.
\end{frame}
%\section{Technology stacks and ethical considerations}
%\begin{frame}
% \frametitle{Key challenges in the study}
%\end{frame}
\begin{frame}
\frametitle{AI \& Ethical Concerns}
This research immediately falls into at least medium risk in Falmouth
University's ethics policy as it involves human participants. Due to the
nature of interacting with generative AI addition measures will be necessary.
\begin{itemize}
\item We should avoid giving AI access to children and vulnerable people.
\item We should make sure to use ethically trained AI models where possible.
\item Furthermore, we should follow CONSORT-AI \cite{Liu2020} to keep this
research transparent and reproducible.
\end{itemize}
\end{frame}
%\section{Me, My Skills, And why I want to study at Falmouth University}
% 1 4 9? 10?
\begin{frame}
\frametitle{Who am I?: Relevant Job Experience}
\begin{itemize}
\item Senior Technician at Falmouth Universities Games Academy, specialising
in Computing.
\begin{itemize}
\item Previously I was an Associate Lecturer and an e-Learning Developer
\end{itemize}
\item I have worked on streaming interactive 3D Architectural Visualisation
experiences with \href{https://www.amutri.com/}{Amutri Ltd}
\cite{AmutriLtd2025}. \item And I have worked with live audio streaming for
podcasts in a former startup called Ramble that attempted to live stream
podcasts and call in radio shows.
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Who am I?: Relevant Research}
I have been contributed to the following papers with academic staff from
Falmouth University previously:
\begin{itemize}
\item Student Perspectives on the Purpose of Peer Evaluation During Group
Game Development Projects \cite{Mitchell2021}.
\item An Exploratory Analysis of Student Experiences with Peer Evaluation in
Group Game Development Projects \cite{Mitchell2022}.
\end{itemize}
\end{frame}
\begin{frame}
\frametitle{Why I want to study here}
\begin{itemize}
\item I already work here delivering content to the students and feel that
the computing departments research goals and my work already align really
well.
\item The research area I'm applying to perform research within is the work
I am already performing at this institution.
\item It will be a good reason to continue to develop new software keeping
up with novel techniques which can also influence my teaching.
\end{itemize}
\end{frame}
\begin{frame}[allowframebreaks]
\frametitle{References}
{\tiny
\bibliography{references.bib}
}
\end{frame}
\end{document}