Emb(race): Judicial reform. From traffic stops and arrests to sentencing and parole decisions, use technology to better analyze real-world data, provide insights and make recommendations that will drive racial equality and reform across criminal justice and public safety.
Theme:
Problem statement 2 (transparent data about officers)
Brief description of your idea:
The bystander effect describes a behavior pattern where people tend to be indifferent to emergencies when surrounded by like people who are also indifferent. If the test subject perceives themself to be similar to (in other words akin to) other bystanders who are not helping, there is a tendency of the subject to not help either. This can be tested in a controlled environment by presenting an unknowing subject with a fake emergency and a set of like and unlike other bystanders who are or are not helping. If the subject feels a connection to the peer bystanders who are helping the subject tends to help. If peer bystanders are not helping the subject tends to not help. Using racially mixed sets of actors in the test may be able to differentiate candidates who are biased vs not biased by detecting which actors they consider peers and which they do not consider peers.
It is unknown whether a virtual environment would be adequate or if a physical experience is necessary. If virtual environments are adequate then the idea can be implemented via technology alone. Even if a physical experience is required, technology can help organize and score the test.
What makes your idea unique?:
If the idea works, it has the potential to empirically determine bias in existing or candidate officers in a safe environment.
What would be the impact of your idea if implemented?:
Reduction in bias in the officer population. Potential for training uses.
Skills to contribute (e.g. development, architecture, research, design or anything else):
For the pure virtual approach needs 3d modeling skills.
Behavioral researchers
UI, database, combinatorial test design development
Theme: Problem statement 2 (transparent data about officers)
Brief description of your idea: The bystander effect describes a behavior pattern where people tend to be indifferent to emergencies when surrounded by like people who are also indifferent. If the test subject perceives themself to be similar to (in other words akin to) other bystanders who are not helping, there is a tendency of the subject to not help either. This can be tested in a controlled environment by presenting an unknowing subject with a fake emergency and a set of like and unlike other bystanders who are or are not helping. If the subject feels a connection to the peer bystanders who are helping the subject tends to help. If peer bystanders are not helping the subject tends to not help. Using racially mixed sets of actors in the test may be able to differentiate candidates who are biased vs not biased by detecting which actors they consider peers and which they do not consider peers.
It is unknown whether a virtual environment would be adequate or if a physical experience is necessary. If virtual environments are adequate then the idea can be implemented via technology alone. Even if a physical experience is required, technology can help organize and score the test.
What makes your idea unique?: If the idea works, it has the potential to empirically determine bias in existing or candidate officers in a safe environment.
What would be the impact of your idea if implemented?: Reduction in bias in the officer population. Potential for training uses.
Skills to contribute (e.g. development, architecture, research, design or anything else): For the pure virtual approach needs 3d modeling skills. Behavioral researchers UI, database, combinatorial test design development
News article that prompted the idea https://www.inverse.com/science/study-finds-animals-exhibit-the-bystander-effect