Human Resource Management

Pymetrics: Using neuroscience games to cut Résumé bias

The recruitment process is a costly and time-consuming process and many companies are now turning to gamification as a means of filtering candidates out. Big names such as Marriott, and the US Army have used games to assess candidates. We speak with Frida Polli, Pymetrics co-founder about the potential of neuroscience games in recruitment, Résumé bias, and her thoughts on the critics.

What inspired you to come up with Pymetrics?

I am an academic neuroscientist and had done my PhD and postdoc at MIT when I realised it wasn’t the greatest fit for me. I spent a long time working towards this goal so I was a little surprised to find that the academic life was not what I thought it was going to be.

So I started to think about other ways I could apply my science to other problems. I ended up applying and going to business school and it was at business school that I saw what was being offered to students such as questionnaires and focus groups: what we would consider as subjective measures of assessment. No one had really thought about applying more objective, scientifically based measures. And that’s how the idea for Pymetrics came about.

Tell me about the games and the match-making process

People looking for jobs [use our games] for career assessment and if [the person] is found to be a good fit and the algorithm matches with that company we will connect the person with the company. It’s exactly like online dating but for companies and candidates and we assess both the candidate’s side and the company’s side using the games and if their profile fits we put them together.

 What do your games assess?

They assess very standard features that we assess all the time in neuroscience. So we look at cognitive traits like memory, attention-span, planning, sequencing, things like that. Different cognitive functions.  We also look at emotional components of behaviour such as risk-preference, reward-preference, and ability to tolerate ambiguity - very standard traits that are very commonly used in research and healthcare contexts.

How did you choose these particular games?

My co-founder Julie Yoo is also a neuroscientist and I met her at MIT. We interviewed a lot of people in the recruiting space to find out what they wanted to assess and then we built a platform of 20 games. We then worked with specific companies trying to profile different roles and then we narrowed it down to 12 based on the fact that these games probably gave 85 – 90% of the information that we needed. So that’s how we selected them.

Were the games developed by you?

We didn’t develop [the games ourselves] but they are used all over the globe currently by neuroscientists. It’s an open source library so to speak and we just borrowed them. We think this is a huge advantage because they are commonly accepted as industry standards - so if I put this game in front of any neuroscientist in any country and say: this is a game that measures attention – it’s very unlikely that they wouldn’t use it because it’s been around for so long and that’s what it measures.

What kind of profile do you build?

Both personality and cognitive based.  If you play all 12 games we measure 50 plus traits and we feedback a lot of the information to users so that they can understand their strengths and weaknesses. Every game will populate different results.

None of the games are supposed to be inherently good or bad  - some people hear cognitive and think it’s an IQ test – it’s not an IQ test  - it’s more about understanding your general cognitive style  - like trying to figure out what kind of athlete someone is  - are you a basketball player or a tennis player? Rather than saying are you a good athlete or a bad athlete.

There are already a lot of personality tests out there. What makes you unique?

Most of the personality tests out there use questionnaires so that’s what probably differentiates us from 90-95% of things that are out there. I think the problem with any questionnaire is that you may not be the best judge of yourself. You may be too harsh at times and not harsh enough. Very few out there do what we do which is gather information about someone using games and look at those cognitive and personality traits. It is important to look at both and not just one or the other.

Some people say you shouldn’t use games to assess a person’s ability.  What would you say to that?

These games are not like your typical video game at all – we call them games because that’s probably the best use of the word. Really they are just activities that you are doing on a computer - there is no benefit to being an avid video-game player versus not being one.

Another thing is that there is no right or wrong. If you score in one direction that means you are better suited for a certain job versus another. The problem is the misperception that all the assessment tools assume there is a good and bad direction – that is definitely not the case – it’s all about fit.

The other criticism we get is that these games are biased against certain ethnic groups or gender and there are two things I would say about that. For one, we never use an algorithm unless it’s unbiased. Plus we are a company started by two women so I’m not sure why we would build algorithms which discriminated against gender!  Second thing to say is that the current [recruitment] process is biased.  Résumé review is by human unconscious bias – so our algorithms are used by companies to get rid of bias rather than include it.


« What's your knack? Games to "measure" job potential


C-suite career advice: Davor Dubokovic, CloudSense »
Ayesha Salim

Ayesha Salim is Staff Writer at IDG Connect

  • twt
  • Mail


Do you think your smartphone is making you a workaholic?