Polina Borisova

Working Papers

"Self-Prospecting: Optimal Experimentation Under Present-Bias" (with Nikhil Vellodi), SSRN.

A present-biased decision maker (DM) faces a two-armed bandit problem whose risky arm generates random payoffs at exponentially distributed times. Under full information, the DM's belief remains unchanged prior to payoff arrivals, generating a "lumpy'' belief process that updates infrequently but conclusively. Our main finding is that coarsening the DM's information to foster "gradual optimism''—a continuously increasing path of beliefs during active experimentation—helps motivate the DM more effectively and deliver them greater welfare. We relate our findings to those in behavioral psychology relating to motivation, learning, and self-control, and apply our results to parenting and pedagogy. 

"TikToks vs. Movies: How Content Length Shapes Engagement" (current draft)

A decision maker chooses between an outside option and consuming content from a creator. Consumption generates a sequence of videos whose quality is stochastic and depends on that latent characteristic. Thus, observing content provides information about both the current video and the creator. Videos end stochastically, and restarting entails a fixed cost. 

Shorter videos accelerate learning about the creator, while reducing the gains from continued consumption of individual high-quality videos. When the engagement cost is high, faster turnover encourages experimentation; when the cost is low, longer content is optimal. This mechanism rationalizes systematic differences in content length across platforms and, in an extension, implies a short-to-long funnel: short formats are best for discovery, while long formats become optimal after adoption.

Work in Progress

"Goals That Motivate, Goals That Inform" 

I study a dynamic principal-agent model in which goals simultaneously function as motivational reference points and as Bayesian screening devices. The principal optimally sets goals to trade off incentive provision against learning about the agent’s type, implying a novel dynamic pattern: after an unmet goal, it can be optimal to raise subsequent goals rather than relax them. 

Other Publications(Mathematics)

"Separation of Variables for Type D_n Hitchin Systems on Hyperelliptic Curves"

Russian Mathematical Surveys 76(2), 181–182 (2021). https://doi.org/10.4213/rm9935, ArXiv.

"Hitchin Systems on Hyperelliptic Curves" (with Oleg Sheinman)

Proc. Steklov Inst. Math. 311, 22–35 (2020). ArXiv.