How do you prepare for the University of Toronto MPI interview?

Toronto MPI interview preparation differs from MMI preparation because the Modified Personal Interview is four independent personal interviews, each assessed by a different interviewer, and each approximately 12 minutes long in person. The virtual version keeps the four interviews but makes them asynchronous: 30 seconds to read the question, a 3 second countdown, then 5 minutes of recorded video, inside a circuit of five timed stations capped at 30 minutes total. So you prepare four separate first impressions, not a scenario circuit, and you rehearse talking to a lens for five minutes with nobody reacting.

What is an MPI and how is it different from an MMI?

Toronto developed the MPI itself and launched it in 2014. The distinction that matters: an MMI station hands you a situation and marks how you reason through it. An MPI interview is a personal interview. Twelve minutes with one interviewer whose job is to form an independent impression of you, your motivations, your file and your judgment.

Toronto is the only Canadian medical school using this format, which has two consequences. First, generic MMI preparation transfers only partly. Second, most prep material you will find online is built for the MMI, so you have to adapt it yourself or work with someone who knows the difference.

Interviewers may include physicians, medical students, residents, other health professionals and community members. Four different people, four different professional vantage points, no shared notes.

What does “four independent interviews” actually change?

Three things.

You get four fresh starts and four fresh judgments. Nothing carries. A story that worked in the first interview is new to the second interviewer, so reuse is allowed. Equally, a strong first interview earns you nothing in the fourth.

Consistency is being sampled, not rewarded directly. Four assessors independently rating the same candidate is a reliability design. Your job is to be recognisably the same person in all four, which means your reasons for medicine have to be true enough to say four different ways.

Your worst interview is a quarter of your score, not a footnote. In a ten-station circuit one bad station is ten per cent. Here it is twenty-five.

How do you prepare for the asynchronous version specifically?

The vMPI is the version currently in use, and the timings are tight and specific. Rehearse against them exactly.

• 30 seconds to read. Practise reading a prompt in 30 seconds and extracting the actual question, not the topic. Read it twice if it is short.
• 5 minutes, used fully. The single most common vMPI failure is finishing in two minutes. Build a five-minute shape: the direct answer, then the specific example, then what you learned or what you would do differently, then a clean close.
• No listener. Record and watch. Without a nodding human, candidates drift into either a presentation voice or a trail-off. Both are visible on playback and invisible in the moment.
• A clean ending. Say the last sentence you mean to say and stop. Do not fill the last twenty seconds with “so, yeah”.
• The ID station. 90 seconds, front and back of government photo ID. Have it physically in hand before you start, not in a drawer.

Compare with a circuit school to see why the preparation differs: UBC states 10 stations and roughly 10 interviewers, which is a stamina problem. Toronto is four deep conversations, which is a consistency problem.

An MPI-style question, answered badly and well

An original prompt, written for this page and not taken from any interview:

Tell us about a time you changed your mind about something important.

Weak answer, and it is weak in a specifically MPI way. “I used to think that being a good doctor was mostly about knowledge, but through my volunteering I realised that communication and empathy are just as important. This changed my perspective and now I try to listen more than I speak.”

That is ninety seconds of a five-minute slot, it contains no event, and the change of mind is one every applicant claims. An interviewer with twelve minutes will ask what specifically changed it, and there is no answer underneath.

Strong answer. “For about two years I argued, genuinely argued, that patients who missed appointments at the clinic I volunteered at should be dropped after three no-shows. My reasoning was that the slots were scarce and someone else could have used them, and I thought people who cared would turn up.

What changed it was a single afternoon of phone calls. We were clearing a backlog and I was asked to ring the no-show list to rebook people. I made about thirty calls. Of the people I reached, the reasons were almost never what I had assumed. Two had no transport because a bus route had been cut. One had been in hospital. One had dementia and had not understood the letter. One had come on the right date the previous month and been turned away because we had double-booked, and had decided we were not serious.

The part that actually changed my mind was the last one, because the system had failed first and then blamed the patient for it. I had been reasoning from a model of people who choose, and the list was mostly people who could not.

What I did with it: I raised it with the clinic coordinator and suggested we ring people two days before rather than send a letter, which we started doing for the following month. I do not know the exact effect, we were not measuring properly, but the room was noticeably fuller.

And what I took from it that I still use: when I find myself explaining someone’s behaviour by their character, that is the moment to go and ask them. It is usually cheaper than I expect and I am usually wrong.“

Why it scores. It names a position the candidate actually held and that is unflattering, which is the hardest and most credible thing to do in a personal interview. It identifies the specific event that changed it, with detail that cannot be fabricated. It states what was done with the insight and admits the outcome was not measured, which is honest rather than inflated. It closes with a transferable rule. And it fills five minutes without padding, which is the format’s real constraint.

Key numbers

• 4 independent interviews, 4 different assessors, in both MPI and vMPI.
• 12 minutes, approximately, per in-person interview.
• 5 minutes of recorded response per virtual interview.
• 30 seconds to read each question, 3 second countdown, 90 seconds at the ID station.
• 5 timed stations and a 30 minute maximum for the whole virtual circuit.
• Approximately 750 applicants invited for the 2025-2026 year.
• 25%: how much of your result one weak interview represents.

Rehearse four five-minute recordings and have them marked: book an interview session.

Related: Panel interview vs MMI and How to answer “tell me about yourself” in a medical interview.

Last reviewed 17 September 2026. Written by Arvin Ardakani, final-year medical student, 4th quartile on CASPer four times.