Monthly Archives: July 2016


One of the best parts of interning at ArticuLab this summer been the diversity of the staff both in culture and disciplines. My fellow interns come from all over: India, China, France, Texas. I am one of the few summer interns that is from Pennsylvania! This diversity from where we come from has allowed for a continuous flow of unique and different ideas and thought processes. I never thought when coming to this internship I would learn as much as I have about different cultures. From going out to dinner to exploring nearby parks, I have learned just as much from the conversations in the lab to the ones outside work.

On top of this mixing of so many different cultures, the interdisciplinary of the research being done provides further to the diversity. My favorite part of Dickinson is the interdisciplinary approach to education. I am able to take classes based on all my interests from computer science to environmental studies. Working in an actual lab that fully supports and encourages interdisciplinary research has been amazing. I have gotten to interact with professionals in Computer Science, Machine Learning, Linguistics, Psychology, and Human-Computer Interactions. Working in this lab has revealed further the power of a liberal arts education.

My time at ArticuLab is quickly coming to a close. I only have two weeks of work left! Thanks again to the Dickinson Internship Grant for providing me the funds to be able to pursue this experience!

To close out this post, here is a picture of me and my fellow ArticuLab Summer Interns!



Inspiration from The Wizard of Oz

One of the great unexpected twists from the classic movie, The Wizard of Oz, as a child was the revealing of the Wizard not being some magnificent god. Instead he was just a plumped, normal human. This deceiving act is the inspiration of an experimental set up used in human-computer interaction research: A Wizard of Oz Experiment or WoZ for short.

A difficult aspect in this field’s research is the time-intensive and expensive nature of creating and programming new software and algorithms.  Many researchers in the field start with a hypothesis such as: a virtual assistant that can develop rapport with a human is more effective in aiding humans with their tasks than a a virtual assistant that does not develop rapport with a human such as Siri, Cortana, Google Now etc. To best verify this hypothesis, a virtual assistant would be built that can develop rapport. Then this rapport-building virtual assistant would be tested on humans to see if it more beneficial compared to a current virtual assistant by using a measured and controlled test.

This process seems straight forward until lets say that the tests reveal that the rapport building virtual assistant is not more effective in aiding humans with tasks than the normal virtual assistant. Now all this time, energy, and money has gone to waste into developing this brand new, never done before virtual assistance that is not more useful than older technology.

This is why WoZs are use! A WoZ experiment starts with telling a human that a computer program is fully autonomous while in reality a human in another room is controlling what the computer is doing. So in this example, instead of creating a automated rapport and dialogue selection system, all that is created is a user interface that a human called a Wizard can control to make it seem that this virtual assistant that is developing rapport is fully autonomous. This way is much cheaper and quicker than the prior description. If the hypothesis gets verified, then the actual automated computer software is created. A WoZ can be used throughout the design process by slowly getting rid of the scope of control that the Wizard has over the virtual assistant until the Wizard is no longer needed!

It is such a simple concept, but the benefits of it are huge! Instead of starting right away building technology that might not be useful, a person first tests a human (Wizard) controlled scenario that mimics what the computer would do without the subject knowing. If it proves effective, then the actual, autonomous software is programmed and created!

My main job at ArticuLab has been researching past and present WoZ design philosophy/systems and creating a WoZ interface to use on a lab project. Having the ability to learn and experiment with different techniques such as utilizing a WoZ has really helped me understand the wide variety of ways researchers conduct their research.


*** ArticuLab’s work on this hypothesis can be found here: SARA

Mastering the Elevator Pitch

The project I am apart of at ArticuLab this summer is the Rapport-Aligning Peer Tutoring (RAPT) project. This week in our staff meeting we had to prepare and present an “elevator pitch” to everyone in the lab. An elevator pitch is an under 30 second speech trying to sell your research.

So the scenario goes: You are at a research conference and lets say you enter an elevator with Bill Gates. You have only five floors to make your case to him on why your research is so important and to spark his interest. This interest should hopefully lead to funding for your research.

With elevator speeches, you are trying to sell your research. Therefore, a great source for inspiration are infomercials. An elevator pitch should not be wordy or complicated by tons of technical terms. Instead, it should spark interest in the listener. As a researcher, funding is crucial to any project and the more you understand how to sell your work in a manner that reveals its importance to a person outside the field, the more likely you will receive the funds you need.

Below is a small example of an elevator pitch I came up with for the RAPT – the project I am working on:

Education disparities between the poor and rich is a real problem. Tutoring should be an avenue to help close this gap, however, the cost of tutoring have made it an avenue only for the rich. What if I told you we could provide highly-effective, low cost, maybe even free, tutoring to all? My research focuses on creating a virtual tutor that can develop meaningful conversation with its student. Meaningful conversation or rapport is crucial to effective tutoring. It is our goal to create an effective, rapport-building virtual tutor which could act as a key to helping to eliminate educational disparities.