Guiding doctoral students through research struggles

The doctoral journey is often romanticised as a solitary quest for knowledge. In reality, this emotional and psychological marathon is a transformation from student into resilient and independent researcher. As supervisors, our role extends far beyond that of a subject matter expert. We are architects of research environments, cultivators of intellectual courage and anchors during periods of doubt. 

One experience early in my supervisory career profoundly shaped my philosophy in this area. A PhD student who had excelled in every course with apparent ease came to my office and announced his intention to quit the programme. I was shocked. He was brilliant and articulate, and had a deep grasp of cybersecurity fundamentals. But the light in his eyes had dimmed. The culprit was the brutal reality of academic publishing. His papers had faced multiple rejections, and the confidence he had built in coursework was crumbling in the face of this new, ambiguous challenge. His struggle was a stark reminder that the core of a PhD programme is not the consumption of knowledge but its creation, which is inherently uncertain, iterative and publicly scrutinised. 

Reflecting on that student and many others over the years, I have developed a supervisory framework built around four principles:

• align research with innate strengths
• design collaborative ecosystems rather than isolated projects
• encourage bold, first-of-its-kind enquiry
• embrace multidisciplinary thinking when progress stalls.

Together, these principles transform the research struggle into intellectual maturation.

Identify and cultivate innate strengths through early exposure

The strengths with which students enter doctoral programmes are often unarticulated. Some think in abstractions and thrive on formal proofs. Others are natural system builders. Still others possess exceptional experimental discipline and data intuition.

The result, if these are not evident, can be misalignment between a student’s cognitive strengths and their research direction, which risks silently eroding their confidence. What initially appears to be lack of ability is often simply a lack of fit. So, a one-size-fits-all research model is a recipe for disengagement and frustration.

• Why professional development in graduate supervision matters
• How to design and supervise GenAI-integrated doctoral research
• An octopus with three hearts: how to approach PhD supervision

To address this, I involve junior students in multiple “research tasters”. These are short, bounded projects that expose students to different methodological paradigms: formal verification, prototype construction, empirical experimentation, user studies or measurement work.

This process is not just about finding a research topic; it is about self-discovery. Rather than prematurely locking students into a dissertation topic, we help them to see where their intellectual instincts and curiosity align, building a foundation of confidence and genuine interest before the high-stakes publication cycle begins.

Build a supportive and collaborative environment

The myth of the lone genius is profoundly misleading, and in modern research it is also dangerous. The most impactful work is rarely done in isolation. As supervisors, we have a responsibility to build an ecosystem where students can thrive. This means securing funding and infrastructure to provide resources, but more importantly it means fostering a culture where collaboration is expected and valued.

A doctoral student in my group, for instance, was working on secure access and authentication mechanisms for multimedia content. She possessed deep expertise in cryptographic protocols but limited knowledge of multimedia systems: video codecs, compression artefacts and content delivery networks. Rather than allowing her to struggle in isolation, I facilitated a structured collaboration with a research staff member specialising in multimedia systems. By working together, they bridged a critical gap. The student provided the security rigour and her collaborator provided the domain-specific context to make the work relevant and novel. The outcome was transformative, not only in terms of high-quality publications at top venues, but in mindset. The student learned that successful research is often a team effort, and that leveraging the expertise of others is a strength, not a weakness.

Encourage bold questions and first-of-their-kind discoveries 

Sometimes, students struggle because they are working on problems that are safe but uninspiring. Passion often emerges when students recognise that their work addresses consequential, real-world questions.

In 2012, I encouraged my students to investigate the security of Apple’s iOS platform. For a platform used by hundreds of millions, soon to be billions, any cracks in this armour would have global impact. My students’ work demonstrated multiple proof-of-concept attacks, including pass-code cracking and interfering with telephony, that substantially challenged the prevailing narrative that the platform was largely impervious to malware. When Apple patched these vulnerabilities, it was not just a validation of their findings. It was a lesson in how rigorous, curiosity-driven research can shape the real world.

Embrace multidisciplinary thinking to break intellectual deadlocks 

When students become stuck, the solution often lies in looking outside the traditional boundaries of the field. Encouraging a multidisciplinary perspective can unlock new questions and provide novel methodologies to answer old ones.

One of my doctoral students struggled with leakage-resilient password systems. For decades, researchers had tried to design password-based authentication systems robust against observation without trusted hardware. The literature was filled with partial constructions and subsequent breaks. After numerous iterations, progress stalled. We sat down and reframed the problem. Instead of asking: “How can we make this more secure?”, we asked: “What cognitive burden does this impose on users?” 

Tapping into cognitive psychology, the student incorporated human cognition into his security research. He broke the authentication process into atomic cognitive operations, which allowed him to demonstrate an inherent trade-off: a secure, leakage-resilient password system inevitably imposes a significant cognitive workload. This finding suggested that the longstanding difficulty in designing such systems was not merely technical, it was human. This multidisciplinary pivot transformed frustration into breakthrough, and the work received a distinguished paper award at a top cybersecurity conference. 

The inevitable nonlinear path of research

So, when a doctoral student says: “I am struggling”, it is not a failure of training. It is an inflection point of growth. The PhD path is inherently nonlinear. It is paved with rejection, ambiguity and false starts. The supervisor’s responsibility is not to eliminate these challenges but to contextualise them. By aligning research with strengths, cultivating collaborative ecosystems, encouraging bold and consequential questions and promoting multidisciplinary openness, we can help our students not only overcome their struggles but also emerge as the resilient, creative and independent scholars the world needs.

Robert Deng is a professor of computer science at Singapore Management University.

If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.