A consortium of researchers dedicated to improving the understanding of the human causes and consequences of terrorism

Unconventional Weapons and Technology team present research on influence operations at Laboratory for Analytic Sciences symposium

Earlier this month, Unconventional Weapons and Technology (UWT) researchers Megan Rutter and Rhyner Washburn and UWT Director Dr. Steve Sin provided a virtual presentation titled “Influence-to-Action Model (I-AM) Development: Measuring Foreign Influence Operation Impact” at the 2021 Laboratory for Analytic Sciences (LAS) Symposium.

LAS is a mission-oriented translational research lab at North Carolina State University that partners with a variety of academic, industry and government communities to develop new analytic technology and analysis tradecraft. The annual LAS Symposium presents the outcomes of various LAS research projects.

This presentation focused on the progress and results of the Developing Impact and Effectiveness Assessment Tool for Influence Operations project, which seeks to create a model for measuring the impact and effectiveness of foreign influence operations when it comes to audience behavior change.

In the presentation, Sin explained that the objective of the project was to find out if the team could develop a social science model that would allow them to assess whether influence operations were successful in getting people to change their behavior.

“To accomplish this, we first consulted the extant literature in the cognitive sciences, sociology, political science, marketing, communications, information sciences, terrorism and radicalization, and cybersecurity to determine a set of variables that have been found to be relevant,” Sin said.

This thorough literature review of multiple disciplines yielded some common variables, and when the team focused on those variables relevant to a person at the individual and societal levels, they discovered that such factors appear in influence operation narratives.

The individual factors included emotional appeals, threats to the individual’s identity, threats to a way of life or culture, reaffirmations of a person’s identity, victim affirmations and victimization. The societal factors included a sense of persistent and increasing threats, a clear alternative often involving a better future, a purported lack of a response from anyone else, clearly identified in-/out groups and social validation.

“Once we decided on the factors and established the model, we performed three case studies, of one historical and two ongoing foreign influence operations, to validate the model,” Sin said. “We were successful in applying the model to analyze the three cases, and found that most of the variables were applicable across all operations. The model was also especially useful in discerning similarities across time for influence operations, and noting potential for where messaging may circulate.”

The team found that the more factors from the model that were present in the influence operation, the more successful the messaging was in terms of being spread across the community receiving that messaging.

“Crafting a campaign which is clearly targeting a specific audience that utilizes messaging that contains the factors identified in the model and places them in the target audience’s cultural context had a higher impact than those campaigns targeting broader, or multiple audiences at once,” Sin said.

The team also created a survey to identify which individual and societal factors are most important to the U.S. general public, asking participants what types of messages would make them want to share, respond or act, such as making financial contributions or volunteering time.

“Overall, 33 percent of respondents answered that they would act if the messaging makes them feel affirmed about who they are and what they represent,” Sin said. “Additionally, 30 percent of the participants responded that they would act if the messaging makes them feel they are doing something to make the world a better place.” Sin noted that the margin of error for the survey was +/- 3% at 95% confidence level.

Sin explained that the research indicated that the more narrowly focused the influence operation, the higher impact it tends to have, and that people tend to respond to messaging that contains identity affirmation and a clear, better alternative future, more than any other types of messaging.

“An interesting and important finding here was that the characteristics and ideology of an individual do not seem to provide us with any useful distinction if one would act based on one type of message versus another. But in general, people tend to act on messages that provide them identity affirmation and a clear alternative or better future,” Sin said.

The project team has finished up the first year of the study, and the funder has since extended the project end date so that the team can conduct a second round of surveys to obtain a more nuanced understanding of their findings, which will enable them to develop a better measurement metric as well as the Influence Impact Assessment Tool.

Sin closed the presentation with a recommendation on how to counter such influence operations.

“Given our findings, we suggest the possibility that messages conveying identify affirmation and a clear, alternative future could prove to be fruitful counter-narrative approaches to combat mis- and disinformation campaigns,” Sin said.