Television news in the last several weeks has seemed eerily reminiscent of the period after Sept. 11 2001, flooding the airwaves with frightening, violent imagery, and a sense that terrorism is likely. According to the media monitoring service TVEyes, CNN mentioned the Islamic State more than 3,800 times in the past several weeks. The Islamic State itself utilizes the media to spread its message and recruit followers  through gruesome videos of attacks and beheadings, creating fear in publics targeted for attacks. Americans are paying attention to this news and are increasingly concerned about the Islamic State. In a recent Wall Street Journal/NBC poll, 94 percent of respondents were familiar with the murders of two U.S. journalists by ISIS, and 47 percent think that the country is less safe than before the 9/11 attacks. Again, just like the period after Sept. 11, Americans seem to support military action in the Middle East. They are increasingly in favor of military strikes against the Islamic State. In a Sept. 20 Gallup Poll, 60 percent of respondents supported air strikes, up from 39 percent support in June, at the beginning of Islamic state sieges on Mosul, Tikrit, and Sinjar.

Political and media observers, particularly on the left, worry that media coverage of the Islamic State is terrifying Americans and persuading them to support foreign policies and candidates that they would otherwise not support. Political science suggests that their fears are warranted. My own research – conducted in the wake of 9/11 – provides strong evidence that both the amount and tone of media coverage of terrorism can significantly influence foreign policy attitudes. Americans who were already worried about future terrorism after 9/11, were more likely to support the use of military force abroad and increased spending on security at home after seeing news stories about terrorism with images like the World Trade Center on fire.

My research used two statistical approaches (OLS regression and an instrumental variables approach) to analyze data from an American National Election Studies 2000-2002-2004 panel that interviewed the same respondents both before and after the 9/11 attacks, asking how TV viewing changed public attitudes to security policy. TV news viewing didn’t much change the views of respondents who thought that future terrorist attacks in the United States were unlikely in the short term. However, as the graphs below show it did appear to change the views of the two-thirds of respondents who thought that future attacks were likely or very likely. The more TV they watched, the more hawkish were their views. Democrats were especially affected by televised images of terrorism. Democrats who had watched the World Trade Center fall were more likely to have warm feelings toward President George W. Bush than Democrats who had not seen this image .

Figure 1: TV news exposure increases support for hawkishness for Americans concerned about terrorism

Even five years after 9/11, seeing televised images of terrorism made people more likely to support the types of hawkish counterterrorism policies advocated by the Bush administration. In 2006, I worked together with YouGov/Polimetrix to create an experiment involving a representative sample of 1,220 Americans. Some of these 1,220 were picked at random to watch a TV story unrelated to terrorism, some to watch a TV story about terrorism with neutral images and no emotional commentary, and some a TV story about terrorism with scary visuals chosen to emphasize the threat of terrorism, such as the burning World Trade Center and bloodied victims of the 2005 London transit bombings. After watching one of the three news stories, respondents answered a series of questions on U.S. foreign policy.

The graphic below shows that scary visuals have a big effect on people who think that terrorism is likely or very likely in the near future. The black circles represent the differences in attitudes between respondents who saw the neutral story about terrorism and respondents who saw the story that was unrelated to terrorism. This tells us about the consequences of watching TV news items with threatening information about terrorism. The white open diamonds represent the differences in attitudes between people who saw the TV news item with scary visuals about terrorism, and people who saw the TV news item with neutral visuals about terrorism. This tells about the consequences of watching TV news items with frightening imagery. When people’s foreign policy views move in a hawkish direction, they are shown on the right of the vertical dotted line and when they move toward the dovish end appear to the left of the line.

Not only did the scary visual condition increase support for hawkish policy by about 6 percent compared to the news story which did not talk about terrorism, but the TV news item with scary visuals increased support for militaristic policies more than the TV news item with neutral visuals, even though they provided the exact same information. Respondents who saw the news item with scary visuals were more likely than those who watched the version with non-emotional imagery to support military solutions to international problems and higher spending on areas such as defense and border security, and to have more favorable views on the government’s handling of terrorism. Although not all of the differences between the neutral and scary visuals conditions reach conventional levels of statistical significance, the overall pattern of findings is very clear. Threatening news makes people more hawkish even on its own, but it is most influential when it is presented in a sensationalistic way.

Figure 2: Scary terrorism imagery increases support for hawkish policy

Terrorism is newsworthy because it is inherently dramatic and threatening. Media competition means that journalists and editors have incentives to use emotionally powerful visuals and story lines to gain and maintain ever-shrinking news audiences. During the years after 9/11, this sensationalistic news coverage increased support for the hawkish foreign policy advocated by the president.

For sure, graphic news imagery does not always move the public toward hawkishness. Disturbing images of chemical attacks on Syrian civilians were not enough to convince a majority of Americans to support military action there in 2013. Where media images of terrorism enhance a sense of fear and political leaders agree on a set of policies to counter threats, the American public is likely to support those policies. During times of crisis, the political spectrum shrinks, leaving the president as the dominant voice on foreign policy, and making a frightened public more likely to endorse the president’s policies. However, when opposition voices return, Americans are likely to support the foreign policy advocated by their party leaders even when frightened.

All this suggests that frightening media images of the Islamic State will probably make the public more hawkish. The September Gallup Poll tells us that Republicans and Democrats supported air strikes on ISIS at equally high rates (65 percent of Republicans; 64 percent of Democrats), reflecting growing hawkishness among Democrats. Only 34 percent of Democrats supported military strikes in June. As President Obama made the case for military strikes, Democrats in the public, many of whom were probably concerned about future terrorism, adopted the president’s position. Congress approved of President Obama’s plan to arm moderate Syrian rebels in September and approval ratings of the president’s handling of foreign policy are on the rise, again making hawkish policy dominant. No president wants to seem indifferent to or flatly wrong about the risk of terrorism, making it more likely that presidents will see military action as necessary to protect the United States. My research tells us that media coverage of terrorism with searing images and warnings of more violence to come is likely to get Americans who are worried about terrorism to line up behind the president’s use of force.

Shana Gadarian is assistant professor of political science at Syracuse University’s Maxwell School