A computational cognitive modeling approach to understand and design mobile crowdsourcing for campus safety reporting

Abstract

The under-reporting of public safety incidents is a long-standing issue. In this paper, we propose a computational cognitive modeling approach to understand and design a mobile crowdsourcing system for improving campus safety reporting. In particular, we adopt drift-diffusion models (DDMs) from cognitive psychology to investigate the effect of various factors on users' reporting tendency for public safety. Our lab experiment and online study show consistent results on how location context impacts people's reporting decisions. This finding informs the design of a novel location-based nudge mechanism, which is tested in another lab experiment with 84 participants and proved to be effective in changing users' reporting decisions. Our follow-up interview study further suggests that the influence of people's mobility patterns (e.g., expected walking distance) could explain why the nudge design is effective. Our work not only informs the design of mobile crowdsourcing for public safety reporting but also demonstrates the value of applying a computational cognitive modeling approach to address HCI research questions more broadly.

Do I Care Enough? Using a Prosocial Tendencies Measure to Understand Twitter Users Sharing Behavior for Minor Public Safety Incidents

“…Many of our motivations related to social media use behavior are affected by psychological aspects of our personalities. For example, anonymity is thought to have an effect on decisions to share public safety incidents [14], although previous research was not able to find a direct influence of anonymity on reporting decisions [2]. Traits like altruism, for example, were found to highly affect people's tendency to share on social media [20].…”

“…Another study looked at data from an eastern (Korea) and a western (U.S.) country to understand motivations for sharing marketing information on Facebook [21]. In terms of public safety, prior research looked at three different motivations for reporting incidents on campus [14]. Thus, the general act of sharing on social media, whether it is in the form of reposting, retweeting, or actually typing the content, has been frequently investigated.…”

“…Thus, the general act of sharing on social media, whether it is in the form of reposting, retweeting, or actually typing the content, has been frequently investigated. Although decisions to share based on level of crime severity ranging from high to low was investigated in prior work [15], to the best of our knowledge, there is no previous research looking into motivations to share information related to minor public safety incidents on social media.…”

Crisis Informatics in the Context of Social Media Crisis Communication: Theoretical Models, Taxonomy, and Open Issues

“…Crowdsourcing [6], [55], [79]- [81] and digital volunteers [82]- [84] are the additional areas identified from the literature. Furthermore, crisis communication [33]- [35], [38], [85]- [88] is another area of research that is further divided into social network [89], [90], issue arena which involve places for societal discussion on social media [90], [91], others areas include crisis responses in the form of information dissemination [87], [92]- [94] and information seeking [67], [95]- [100]. The effectiveness of social media for information dissemination and sharing has been discussed widely by [101], and [102] studies focus on the information dissemination dynamics, and their influence on emergency management.…”

“…The application of statistical techniques was also used. For example, [110]- [112] used SEM, PLS and Regression analysis in their studies, Anova by [87], [113]. Also, data mining techniques and applications [9], [108], [109] for decision-making purposes were seen in various studies.…”