Join us for a special Wednesday @ NICO with four 10 minute lightning talks from Northwestern University fellows and scholars.

Speakers:

Umit Aslan - Ph.D. candidate, Learning Sciences
Daniel J. Case - Ph.D. Candidate, Department of Physics & Astronomy
Aymeric Punel, Ph.D Candidate, Civil and Environmental Engineering
Orsolya Vasarhelyi, Visiting Pre-Doctoral Fellow, Department of Communication Studies

Abstracts:

Umit Aslan - How do lay people reason about complexity?

The intuitive understanding of complex systems plays a critical role in our lives. Deciding where to move, who to vote for or which career path to follow requires us to reason about systems and processes that are decentralized, unpredictable, and ever evolving. Thus, investigating how lay people reason about complex systems holds the key to designing learning environments and technologies. In this lightning talk, I will present some salient themes that emerged from my studies with first-time agent-based modelers, both adults and high-school students, on how they reason about real-world phenomena (e.g., gentrification, fake news, population decline) and also how they pick up agent-based modeling to investigate complex systems computationally.

Daniel J. Case - Exploiting nonlinear dynamics for programmable behavior in microfluidic networks

Microfluidic networks are a technology widely used across biomedicine, chemistry, and physics for the purpose of precisely manipulating small volumes of fluids (nanoliters). Typically, large external pumps and computers are used to control fluid flows through these networks, which hinders their inherent scalability and portability. Here, I present several microfluidic networks that exhibit an array of nonlinear flow dynamics, such as spontaneous oscillations, switching, bistability, and negative conductance transitions, which enable new mechanisms for built-in, programmable flow control. I show how these behaviors arise from nonlinear fluid mechanical effects that are amplified and harnessed through the design of the network geometry, and thus are not reliant on external control devices. These results are supported by analytic models, rigorous fluid dynamics simulations, and direct experiments.

Aymeric Punel - Modeling Driver Behavior in Crowdsourced Delivery Network

The sharing economy is growing in adoption and relevance. Despite the original promises, it has been shown to reproduce offline biases, and disadvantage already deprived populations. Due to limitations in data acquisition, online crowd-shipping, an emerging and consequential branch of the sharing economy, has received less investigation thus far. We fill this gap by investigating crowd-shipping driver behavior using a unique longitudinal data set from a US leading crowdsourced delivery company. After describing aggregate statistics of regional crowd-shipping networks, we use exponential random graph modeling to examine the significance of attribute similarities between drivers for engaging in similar bidding behavior. Specifically, we uncover that spatial proximity, tenure and performance homophily, as well as similarities between the neighborhoods of drivers are associated with choosing similar shipment requests. Conceptually, these findings help reveal behavioral patterns and map the invisible boundaries to free choice among crowd-shipping participants. Practically, the new knowledge can be used to inform recommender systems that could diversify options and improve experience with the platform.

Orsolya Vasarhelyi - Why diversity is not enough?

Despite improvements, women in STEM are still facing more challenges than their male colleagues: they earn less, have access to less funding, are less likely to be promoted, and their work receives less acknowledgment. These disparities persist, despite evidence that integrated female members increase the overall intelligence of teams and gender heterogeneous teams are more creative and productive.

Live Stream:

bluejeans.com/8474912527

Related Info