Definitions and conceptual issues

The Oxford English Dictionary defines propaganda as “the systematic dissemination of information, especially in a biased or misleading way in order to promote a political cause or point of view.” However, because the concept was not sufficiently comprehensive to describe the full range of influences involved in persuasive activities (in addition to the negative connotations the word acquired from its use by the Nazi and Soviet regimes), it fell out of favor in much of the West in favor of “persuasion,” considered a more comprehensive and less polarizing term (Markova, Reference Markova2008). Still, the two words have a tendency to be used interchangeably, but scholars have proposed an interesting dichotomy to differentiate between them. Propaganda is conceptualized as a one-way “monologic” communication from a “source” to a “receiver,” with the goal of “transform[ing] the heterogeneous thought of individuals into those of a homogenous ‘collective mind’ of masses, and to lead those masses to a specific action” (Markova, Reference Markova2008, p. 41). In other words, propaganda can be thought of as a unidirectional communication in which the source of the message holds the predominance of power with regard to the ability to create, change, or normalize the social reality of the receiver (Markova, Reference Markova2008).

In contrast, persuasion is conceptualized as a two-way “dialogic” communication in which the power between the source of the message and the receiver is more broadly shared, but it is also conceptualized to include one’s internal dialogue and unconscious aspects of thought. However, unlike propaganda, in which the more powerful source seeks to “fuse” its reality with that of the receiver, in persuasion, the source’s aim is to “convince the other party of one’s own case and of the superiority of one’s own idea or belief over that of the [receiver]” (Markova, Reference Markova2008, p. 45). With this in mind, neurocognitive hacking is proposed to support the role of propaganda by making its ingestion more likely and facilitating persuasion by cultivating a neural environment in the receiver more accommodating to the source’s narrative, especially when it involves in-group/out-group dynamics.

For the same reasons discussed earlier, “psychological operations” and “information operations” are terms requiring clear distinctions. Both words are typically used in relation to nation-state-sponsored military or civilian intelligence operations, but they have different operational scopes. The U.S. Department of Defense defines psychological operations (PSYOPs) as “planned operations that convey selected information and indicators to foreign target audiences to influence their emotions, motives, objective reasoning, and ultimately, the behavior of foreign governments, groups and individuals” (U.S. Department of the Army, 2003, p. GL-8).

In modern military operations, propaganda is conceptualized as a tool of PSYOPs. However, because of innovations and advancements in technologies complementary to PSYOPs, along with the efficiencies gained by employing them in concert with other supporting capabilities, Western militaries increasingly refer to information operations as

Additionally, although brain science is considered by many to be in its infancy, thanks to recent developments in brain imaging technology, the field is advancing rapidly and gaining insights into once-invisible processes (Jorgenson et al., Reference Jorgenson, Newsome, Anderson, Bargmann, Brown, Deisseroth and Wingfield2015). As artificial intelligence tools are incorporated into these research efforts, the rate of discovery will likely only increase; however, there are many ongoing foundational debates regarding brain function that have yet to be resolved, such as the relationship between one’s evaluation (i.e., attitude) of an object or concept and subsequent behavior toward it (Ajzen & Cote, Reference Ajzen, Cote, Crano and Prislin2008). As neurocognitive hacking proposes the ability to utilize specific stimuli to activate neural structures (e.g., amygdala and anterior cingulate correlate) for the purposes of exploiting affective reactions and shaping targeted behavioral outcomes, a quick treatment of the mood/attitude–behavior linkage is offered.

Conceptually, regarding the reactions to mortality-related stimuli, the author agrees with the logic of Tritt and colleagues (Reference Tritt, Inzlicht and Harmon-Jones2012) and their use of the term “affect” versus “moods” or “emotions,” as it avoids unnecessary theoretical and sematic “baggage.” For example, there are ongoing debates regarding the extent to which (conscious) cognitive processes are involved in emotion (Winkielman & Berridge, Reference Winkielman and Berridge2004) and attitude formation (Devos, Reference Devos, Crano and Prislin2008). However, by using the construct of “affect” to describe moods or emotions that are inclusive of behavioral phenomenon occurring below conscious awareness (e.g., mortality-related stimuli) and thus opaque to self-reported measures, these controversies are largely avoided.

In fact, research by Winkielman and colleagues (2004) suggests that affective responses intense enough to influence one’s behavior can remain below awareness. Therefore, reactions to mortality-related stimuli (M-rS) will not be conceptualized in terms of an attitude-mediated behavioral construct (e.g., MODE model, theory of planned behavior) but as a biologically construed “mortality bias” composed of unconscious affective reactions (Winkielman et al., Reference Winkielman, Berridge and Wilbarger2005), activated by an energized threat detection system (Tritt et al., Reference Tritt, Inzlicht and Harmon-Jones2012) for the purposes of reducing false negative (Type II) errors (Haselton & Buss, Reference Haselton and Buss2000) and “psychological uncertainty” (Tritt et al., Reference Tritt, Inzlicht and Harmon-Jones2012). A more in-depth discussion of this topic will be presented later in the article.

Psychological operations in conflict

Taylor (Reference Taylor2003) highlights many fascinating uses of propaganda throughout history, covering operations from ancient Greece to the post–Cold War era. His well-regarded book highlights the evolution of propaganda and explicates how its methods generally mirror the overall advancement of technology in a society. In ancient Greece, for example, architectural marvels like the Acropolis were used to persuade citizens and noncitizens alike of the superiority of Greek culture. More than two millennia later during the Cold War, Americans utilized radio (e.g., Voice of America) in their propaganda efforts against the Soviet bloc. As communications technology in the information age continues to advance, its technologies will provide increasingly rich support for disseminating propaganda in new digital formats to match the current crop of polarizing memes, “fake news” on social media, and “deepfake” manipulations of video clips.

Regardless of the polarizing nature of the concept, Taylor suggests that readers approach propaganda as a morally neutral concept best looked upon as a “process for the sowing, germination, and cultivation of ideas” (2003, p. 2). He notes that the Vatican operationalized this process during the Protestant Revolution in Europe to defend itself against heretics with such success that even today, Catholics and Protestants can still be distrustful of one another. This example alone could attest to the incredible power of propaganda; however, Taylor is quick to note that it alone is not enough to win conflicts but is most effective when integrated with other levers of power, such as diplomatic, military, or economic.

Contemporary history is replete with examples of political and military leaders using propaganda and other broader forms of PSYOPs to their benefit—or their peril, if they failed to incorporate it into their repertoire of operational capabilities. For example, John Nagl’s Learning to Eat Soup with a Knife (2002) highlights the positive role PSYOPs played during the 1940s–1950s British fight against the communist insurgency in Malaysia. Nagl points out the British Army after World War II was more of a “learning organization” compared with their American counterparts and more willing to experiment with unconventional tactics, including the use of PSYOPs to win “hearts and minds.” To a large extent, this resulted from the British Army’s experience fighting many “small wars” in remote locations over the previous two centuries and being forced to adapt to myriad enemies and operating environments. However, because the British did not have inexhaustible resources to bring to bear on these limited wars, they focused heavily on understanding (and exploiting) the motivations of their enemies, an effort often made easier by partnering with indigenous civil authorities.

This approach prompted experimentation with various tools of persuasion and influence to help the British exploit and ultimately break the will of their enemies. In Malaysia, for example, British field commanders were given flexibility to employ various PSYOP tactics (including blasting propaganda from loudspeakers mounted on airplanes, placing bounties on the heads of insurgent leaders, and dropping leaflets over enemy territory), an effort that by 1960 largely proved successful (Nagl, Reference Nagl2002).

Nagl highlights how the U.S. Army, in contrast with the British, was less apt to incorporate PSYOPs into military operations as the Americans relied on (and could afford to use) a policy of fielding much larger units, intent on using overwhelming force to exact total destruction on the enemy. He notes that when General William Westmoreland took command in Vietnam, his army was sorely lacking in its understanding of PSYOPs and thus its preparation was “inappropriate to the demands of counterinsurgency warfare in South Vietnam” (Nagl, Reference Nagl2002, p. 174). American deficiencies were compounded by the strength of PSYOPs conducted by their National Liberation Front foes, who, as the scholar Francis Fitzgerald pointed out, were geared toward inculcating the “systematic encouragement of hatred” of the United States. Similarly, having recognized the motivational benefit of generating hatred within the Vietnamese people toward their French occupiers a generation prior, Ho Chi Minh purportedly said, “I have no army, I have no finances, I have no education system, I only have my hatred” (Fitzgerald, Reference Fitzgerald1972, p. 169). Hate is a powerful motivator and the rise of the internet, in combination with other information communication technologies and platforms, has made it easier for both separatists and terrorist groups to generate and exploit it for specific purposes.

One of the more recent and concerning innovations of terrorist organizations is their ability to utilize social media for radicalization and recruitment. Communications scholar Gabriel Weimann (Reference Weimann2015) notes that social media has given terrorists groups an enormous advantage because of the “effectively limitless” audience it creates for recruitment and the ease with which propaganda videos can be uploaded in response to dynamic operating environments. Two cases from nearly a decade ago highlight the speed and effectiveness with which PSYOPs can be used in the cyber domain to support the terrorist radicalization process.

In 2011, Arid Uka, an Albanian Muslim immigrant to Germany, admitted to becoming self-radicalized as a result of consuming online jihadist propaganda. In roughly six months, after watching numerous propaganda videos, including one doctored by the Islamic Movement of Uzbekistan falsely depicting U.S. soldiers sexually assaulting Iraqi and Afghan women, Uka drove to nearby Frankfurt Airport and shot and killed two U.S. military personnel transiting from a base in the United Kingdom (Bohleber & Bohleber, Reference Bohleber and Bohleber2012).

Similarly, the case of British citizen Roshonara Choudhry (Pearson, Reference Pearson2015) highlights a rare instance of a female terrorist attacker being radicalized to action online. In early 2010, Choudhry, a fairly typical university student working toward completing her degree at King’s College, London, admitted to becoming radicalized as a result of spending several months viewing hours of propaganda videos. Many of these videos featured the well-known jihadist Sheikh Abdullah Azzam and the American radical propagandist Anwar al-Awlaki. By the end of the academic year, Choudhry had dropped out of university, become estranged from most of her friends and family, and attempted to stab to death a British member of Parliament. Her case was highly unusual in that most jihadist attacks are committed by men, with the vast majority of Islamic propagandist urging women to participate in support actions only. However, during questioning after her arrest, she noted that she overcame these gender and ideological barriers to action after viewing videos of Sheikh Abdullah Azzam, who decreed “even women” had a duty to engage in jihadist attacks (Windsor, Reference Windsor2018).

Information operations in the digital era

Although the capabilities of terrorist groups to conduct PSYOPs for purposes of recruitment and radicalization present a formidable challenge to global security, they generally lack the resources needed to conduct full-spectrum information operations. The science of information operations in the digital era is advancing rapidly, and it is increasingly characterized by the “integrated employment, during military operations, of information-related capabilities in concert with other lines of operation to influence, disrupt, corrupt, or usurp the decision-making of adversaries and potential adversaries while protecting [one’s] own” (U.S. Department of Defense, 2014, p. ix). Modern information operations apply theories developed from the study of persuasion and motivation, and the capability is becoming increasingly sophisticated as it parallels advances in those fields (Jowett & O’Donnell, Reference Jowett and O’Donnell1986). These advances are supported by new imaging tools such as functional magnetic resonance imaging (fMRI) and co-registration techniques that combine old and new technologies to produce more comprehensive scans. These new imaging tools have illuminated previously unknown neural processes in the brain responsible for interpreting the social world, including highlighting the importance of neural structures such as the amygdala in evaluating stimuli with emotional value. As many of these processes are known to occur below the level of consciousness (Adolphs et al., Reference Adolphs, Tranel, Damasio and Damasio1995) they underscore the susceptibility of neural components like the amygdala to manipulation as their activation may be exploited to shape targeted behavioral outcomes.

In fact, research suggests that subliminal mortality-related stimuli (e.g., an image of a dead body) can activate the amygdala and other neural components associated with the processing of threat stimuli (Quirin et al., Reference Quirin, Loktyushin, Arndt, Küstermann, Lo, Kuhl and Eggert2011). Therefore, exploiting these unconscious processes, combined with the ability to commandeer communication networks supporting smartphones and other electronic media devices, is proposed to offer a powerful tool for increasing the ingestion of propaganda and subsequent shaping of perception and behavior of adversaries. It should be noted that although the effects on behavior from neurocognitive hacking are conceptualized to be small, the ability to “nudge” a small group of people in one direction can have enormous strategic consequences, as highlighted by the fact that the 2016 U.S. presidential election was determined by fewer than 80,000 votes in three states (Bump, Reference Bump2016).

To highlight the growing strategic relevance of information operations in interstate conflict, consider the indictments by the U.S. Office of Special Counsel. It charged Russian agents with interfering in the 2016 presidential election by seeking to “sow discord” within the U.S. political system (U.S. Department of Justice, 2018). As part of their efforts, Russian operatives were accused of creating numerous social media handles geared toward increasing public polarization over several highly contested social issues. According to the indictment, these social media accounts sought to inflame attitudes over issues such as African American and Muslim civil rights, using provocative names, including “Woke Blacks” and “United Muslims of America.” Although it is difficult to calculate the extent to which this propaganda influenced the public, consider if the targeted audiences had first been primed with mortality-related stimuli that prompted the activation of neural correlates related to threat perception and response. It is proposed that such a targeted neurological attack on these neural structures would have increased the rate of propaganda ingestion by the targeted audience and subsequent level of societal discord within the United States. The ability to influence or “hack” the perception of adversaries to shape behavioral outcomes is of increasing interest to national security stakeholders in the United States, many of whom assess the United States’ current capability in this area to be critically underdeveloped (Astorino-Courtois, Reference Astorino-Courtois2017).

Why mortality-related stimuli?

Research utilizing priming stimuli is controversial; however, a meta-analysis by Weingarten et al. (Reference Weingarten, Chen, Mcadams, Yi, Hepler and Albarracin2016) and research by Winkielman et al. (Reference Winkielman, Berridge and Wilbarger2005) suggest they can have a significant effect on behavior. These effects appear even greater when primes incorporate the use of words with negative valence (Nasrallah et al., Reference Nasrallah, Carmel and Lavie2009). In fact, robust research exists suggesting that priming with mortality-related stimuli mediates prejudicial behavior toward out-groups and their respective cultural symbols (Greenberg et al., Reference Greenberg, Pyszczynski, Solomon, Rosenblatt, Veeder, Kirkland and Lyon1990). After perceiving mortality-related stimuli, individuals have been shown to unconsciously exhibit negative biases toward out-groups, including increased hostility and willingness to engage in avoidance behavior, as well as a greater willingness to actively denigrate out-group cultures (Burke et al., Reference Burke, Martens and Faucher2010; Greenberg et al., Reference Greenberg, Pyszczynski, Solomon, Rosenblatt, Veeder, Kirkland and Lyon1990). The preponderance of research utilizing mortality-related primes has traditionally been subsumed under a theoretical construct referred to as terror management theory (TMT).

Motivated by the writings of Ernst Becker, TMT’s theoretical framework is controversial, as it asserts that during the course of evolution, humans reached a level of cognitive sophistication enabling an awareness of the inevitability of death (Greenberg et al., Reference Greenberg, Pyszczynski, Solomon and Baumeister1986). According to TMT, the inexorable nature of death created such maladaptive terror in humans that it prompted the species to generate a suite of psychological coping mechanisms referred to as “cultural anxiety buffers” (Rosenblatt et al., Reference Rosenblatt, Greenberg, Solomon, Pyszczynski and Lyon1989). When mortality-related stimuli are salient in the environment, the buffers are theorized to work by managing the paralyzing effects of death awareness. Therefore, they are purported to facilitate a deeper fusion with one’s cultural worldview to give symbolic immortality and mitigate the finality of biological death (Solomon et al., Reference Solomon, Greenberg, Pyszczynski, Greenberg, Koole and Pyszczynski2004). To test these assertions, TMT researchers utilize a priming manipulation referred to as “mortality salience.”

The effects of mortality salience (i.e., awareness of mortality-related stimuli) are explored in research settings by asking subjects to consider the ramifications of their physical death and examining their subsequent behaviors. In general, TMT researchers have found that priming with mortality-related themes tends to facilitate “worldview defense” (a worldview comprising the foundational beliefs one holds to help understand and interpret the world). A worldview defense, therefore, is roughly defined as actions involving the defense of one’s culturally based belief system. Examples include exhibiting negative discriminatory behavior toward out-groups (i.e., those holding different worldviews) or positive discrimination toward one’s in-groups (Greenberg et al., Reference Greenberg, Pyszczynski, Solomon, Rosenblatt, Veeder, Kirkland and Lyon1990).

Although experiments using mortality salience manipulations to induce prejudiced reactions are numerous (Greenberg et al., Reference Greenberg, Pyszczynski, Solomon, Rosenblatt, Veeder, Kirkland and Lyon1990; Rosenblatt et al., Reference Rosenblatt, Greenberg, Solomon, Pyszczynski and Lyon1989), the framework suggested by TMT has been criticized on several theoretical grounds (Fessler & Navarrete, Reference Fessler and Navarrete2005; Holbrook et al., Reference Holbrook, Sousa and Hahn-Holbrook2011; Kirkpatrick & Navarrete, Reference Kirkpatrick and Navarrete2006). For instance, researchers in the field of coalitional psychology assert that reactions to mortality salience (i.e., mortality-related stimuli) result from a series of prosocial behaviors that evolved in humans to better coordinate social interactions within in-groups, especially behaviors that would be adaptive when reacting to crises or threatening situations (Kirkpatrick & Navarrete, Reference Kirkpatrick and Navarrete2006). Coalitional psychology offers a cogent explanation for why mortality-related stimuli affect social relations, and it, along with other complimentary theories (Haselton and Buss, Reference Haselton and Buss2000; Tritt et al., Reference Tritt, Inzlicht and Harmon-Jones2012), provides a well-grounded explanation for the biological foundations of the phenomenon.

Error management theory, an evolutionary perspective on the development of cognitive biases, proposes that under conditions of uncertainly, reactions to adaptively relevant stimuli (e.g., dead animals, potential threats from out-groups, or sexual opportunities) are biased toward false positive (Type I) errors. For example, men have been shown to overestimate the extent to which women have sexual interest in them, as this bias would likely facilitate greater reproductive success over the long term. Similarly, when primed with fear, individuals presented with neutral faces have been shown to attribute more anger to them (Haselton et al., Reference Haselton, Bryant, Wilke, Frederick, Galperin, Frankenhuis and Moore2009), a bias that is likely to have proved protective during the environment of evolutionary adaptiveness. Given that mortality-related stimuli could resolve into actual existential threats (e.g., finding a dead body in one’s immediate environment could represent the presence of a lethal attacker or a lethal disease), response heuristics encoded to avoid making false negative (Type II) errors would likely have been adaptive.

Regarding the biological mechanics of reactions to mortality-related stimuli, Tritt et al. (Reference Tritt, Inzlicht and Harmon-Jones2012) propose the existence of a “biological anxiety system” activated by states of “psychological uncertainty.” When psychological uncertainty results from a “mismatch” between one’s actual and expected reality, a component of this biological anxiety system, referred to as the behavioral inhibition system (BIS), is triggered. As Tritt and colleagues (Reference Tritt, Inzlicht and Harmon-Jones2012) note, the BIS is thought to be integral to both the function of anxiety and approach-avoidance dynamics, as it activates inhibitory neural components located in the right hemisphere of the brain. This explanation comports with findings by Quirin et al. (Reference Quirin, Loktyushin, Arndt, Küstermann, Lo, Kuhl and Eggert2011) showing activation of the right amygdala and left anterior cingulate cortex (ACC) subsequent to priming with mortality-related stimuli.

Research suggests that subliminal stimuli can trigger affective reactions without conscious awareness (Winkielman & Berridge, Reference Winkielman and Berridge2004); however, there is controversy over this issue. Addressing this, Custers (Reference Custers, Moskowitz and Grant2009) notes that most models of human goal pursuit conceptualize a conscious mechanism for determining whether a goal will be pursued based on the “expected value” of the goal. However, because reactions to subliminal goal primes can occur outside of conscious awareness, Custers asserts that the most logical mechanism for determining the expected value of a goal outside of conscious awareness “would be one that relies on affective [not conscious] processes” (2009, p. 179). Additionally, based on research suggesting that there are differentials in the amplitude of affective reactions to valenced stimuli (Holbrook et al., Reference Holbrook, Sousa and Hahn-Holbrook2011), there are likely only a few affective reactions that are as impactful on behavior as those induced by mortality-relevant stimuli.

With the foregoing theoretical framework in mind, the author proposes the term “mortality bias” (Heslen, Reference Heslen2016) to capture the suite of automatic processes related to reactions to mortality-related stimuli.

Role of the amygdala in shaping perception and behavior

Although several neural structures are thought to influence social cognition and human decision-making in concert with the amygdala (Adolphs, Reference Adolphs2003, Reference Adolphs2009), the structure is of particular interest to the conceptualization of neurocognitive hacking given growing evidence of its involvement in the processing and encoding of emotion, fear, and ambiguity within social contexts (Whalen, Reference Whalen1998).

The amygdala is an almond-shaped structure located deep within the temporal lobes of the brain. Traditionally, it was thought to function solely for evaluating threat-related information; however, it is now assessed to be part of an early vigilance system that works with other neural structures to process the emotional value of stimuli. As such, researchers have surmised that the amygdala may act as a complex area for processing the “social, homeostatic, and survival-related meaning of a class of complex stimuli, such as facial expressions of some emotions” (Adolphs et al., Reference Adolphs, Tranel, Damasio and Damasio1995, p. 5889). This view of the amygdala’s role in facilitating emotional evaluations is supported by other research suggesting that it is more sensitive to negative animate versus negative inanimate stimuli in human social contexts. For example, subjects presented with subliminal images of both threatening animals and threatening inanimate objects, such as pointed guns, experience greater amygdala activation in response to the threatening animals (Fang et al., Reference Fang, Li, Chen and Yang2016). Again, this finding suggests that the amygdala plays an important role with other neural components in linking emotional valence to specific objects.

Combining evidence suggesting that unconscious perception of mortality-related stimuli activates the right amygdala and the left rostral ACC with findings suggesting that an activated amygdala is correlated with the propagation of unconscious racial stereotypes (Phelps et al., Reference Phelps, Oconnor, Cunningham, Funayama, Gatenby, Gore and Banaji2000), a logical leap has been made suggesting that the discriminatory behaviors induced by mortality-related stimuli are mediated by the activation of these neural structures. Therefore, by integrating research suggesting that individuals continuously (and automatically) update their social evaluations of others (Wheeler & Fiske, Reference Wheeler and Fiske2005) with evidence suggesting that the amygdala plays a significant role in facilitating these automatic evaluations (Adolphs et al., Reference Adolphs, Tranel, Damasio and Damasio1995), neurocognitive hacking proposes the ability to exploit this interplay to shape perceptual and behavioral outcomes of targeted audiences.

Additionally, scholarship suggests that subliminal exposure to mortality-related stimuli mediates behavior in a similar manner as conscious (supraliminal) exposures. For example, in one experiment, American participants subliminally primed with the word “death” were more critical of anti-American essays than participants who were subliminally primed with the word “field.” Interestingly, terror management theory researchers have found that reactions (e.g., discrimination toward out-groups) occurring subsequent to conscious mortality primes do not manifest until after a short distraction exercise is given. This is not the case with subliminal mortality-related primes, the effects of which can be immediately observed without a distraction (Arndt et al., Reference Arndt, Greenberg, Pyszczynski and Solomon1997).

The need for a distraction exercise has been proposed by Heslen (Reference Heslen2016) to involve the “System 1” and “System 2” dual-processing cognitive construct suggested by Stanovich and West (Reference Stanovich and West2000) and popularized by Daniel Kahneman in his book Thinking, Fast and Slow (2011). In general, while both systems are believed to work in a complimentary fashion, System 1 is thought to comprise the suite of automatic mechanisms that constantly evaluate and respond to environmental stimuli, whereas System 2 facilitates more conscious, deliberative functions. As such, the distraction exercise following conscious awareness of mortality-related stimuli likely interrupts the conscious appraisals of System 2, thus giving primacy to System 1 dynamics, where automatic behavioral heuristics are generated.

Cognitive aspects of military operations (CAMO)

Several of the United States’ strategic adversaries, including Russia, have been involved in researching the cognitive and psychological aspects of information warfare for decades (Thomas, Reference Thomas2004). However, within the U.S. defense establishment, there is growing recognition that the armed forces are behind in this area of research and lack the ability to incorporate knowledge of the “human/cognitive domain” into military operational planning. As opposed to the physical/kinetic domain (e.g., weapons systems and personnel training), where the United States is considered dominant, its utilization of the human-cognitive domain (i.e., the ability to influence “attitudes and behaviors of populations or opponent forces by manipulating information and otherwise preying on human perceptual vulnerabilities”; Astorino-Courtois, Reference Astorino-Courtois2017, p. 6) is proposed to be lacking. It is a space, however, in which United States’ major adversaries are assessed to have invested heavily. Increasingly, this suite of cognitive tools is conceptualized as the cognitive aspects of military operations (CAMO) and incorporates techniques to exploit three key psychological functions: cognition, affect, and conation (Astorino-Courtois, Reference Astorino-Courtois2017).

In 2016, the Strategic Multilayer Assessment Office of the U.S. Department of Defense published a white paper titled “A Cognitive Capabilities Agenda: A Multi-Step Approach for Closing DoD’s Cognitive Capability Gap” (Astorino-Courtois, Reference Astorino-Courtois2017). The paper made several recommendations for how the United States could close the cognitive capabilities gap proposed to exist with its strategic adversaries, including Russia, China, Iran, and North Korea. In addition to recommending updates to doctrine, it called for the Defense Department to increase funding for “actionable cognitive research” and the development of “analytic tools” to integrate cognitive capabilities with the physical/kinetic aspects of warfighting (Astorino-Courtois, Reference Astorino-Courtois2017). The ability to operationalize these cognitive capabilities in “gray zones” (characterized as intense areas of military competition that fall short of conventional war; see Votel et al., Reference Votel, Clevland, Connett and Irwin2016) is an area in which Western allies, given adequate investment, can significantly increase their effectiveness, in light of recent scientific advancements.

Among the West’s strategic competitors, Russia is assessed to be the most advanced with regard to their research efforts and ability to execute cognitive operations. In fact, during the 1960s and 1970s, Russia developed a theory of information warfare referred to as “reflexive control” to maximize advantages in both cognitive and computer-based decision-making processes. Reflexive control has been defined as a method of deception to relay “specially prepared information to incline [adversaries] to voluntarily make [a] predetermined decision desired by the initiator of the action” (Thomas, Reference Thomas2004, p. 237). Russian military theorists assert that during a conflict, the combatant with the greatest understanding of enemies’ moral, cognitive, and psychological underpinnings, including those of senior decision makers, tend to be most successful because they are better prepared to induce the enemy into making adverse decisions (Thomas, Reference Thomas2004).

The need to increase understanding of the cognitive tools used to manipulate perception was well illustrated by the 2016 U.S. Intelligence Community report on Russian hacking and the subsequent indictment filed by the U.S. Office of Special Counsel in February 2018. The 2016 report, “Background to ‘Assessing Russian Activities and Intentions in Recent U.S. Elections’” claimed that Russia tried to influence the 2016 election to “undermine the U.S.-led liberal democratic order” and “public faith in the U.S. democratic process” (Office of the Director of National Intelligence, 2017, p. ii). The later indictment filed by the Office of Special Counsel highlighted both the organizations involved and the tactics used by Russian operatives to accomplish these goals. For instance, the special counsel charged the Russian Internet Research Agency with election interference in part for its role in creating highly polarized fake social media accounts.

These accounts were geared toward spreading inflammatory and derogatory information for the purposes of creating strife in the U.S. political system. Fake social media profiles such as “Woke Blacks,” “Blacktivist,” and “United Muslims of America” were established in an attempt to suppress the minority vote during the election. Examples of messages on these sites included memes such as, “Hillary Clinton Doesn’t Deserve the Black Vote,” “Hillary is a Satan…,” and “Donald wants to defeat terrorism…Hillary wants to sponsor it” (U.S. Department of Justice, 2018). Given the strong reinforcing dynamics that group membership can exact on individual behavior (Glassner, Reference Glassner and Tajfel1985; Jost et al., Reference Jost, Banaji and Nosek2016), the efficacy of these messages, when engineered to exploit central characteristics of group identity, should not be underestimated (Hogg et al., Reference Hogg, Terry and White1995).

To understand how a constant stream of socially engineered messages can influence one’s emotional state, consider a study conducted by Facebook several years ago. In early 2012, the social media giant initiated an extraordinary experiment aimed at manipulating the emotional states of 700,000 of its newest members. For one week, the site changed these users’ newsfeeds to display a preponderance of either happy/positive or sad/negative news stories for the purposes of assessing any effects on their emotional states. At the end of the week, depending on their respective treatment, users did show a propensity to post either positive or negative words, providing evidence of an “emotional contagion” effect (Meyer, Reference Meyer2014).

Consistent with this research, in 2016, the chief executive officer of Cambridge Analytica—the data analytics firm that assisted multiple African elections as well as, later, the Trump presidential campaign—stated, “If you know the personality of the people you’re targeting, you can nuance your messaging to resonate more effectively with those key audience groups” (Nyabola, Reference Nyabola2019, p. 1). A former employee of Cambridge Analytica turned whistle-blower offered a more succinct and colloquial analysis of his company’s mission, noting that the company had succeeded in developing a “psychological warfare mind-fuck tool” (Halpern, Reference Halpern2018).

In fact, during testimony to the U.K. Parliament, the employee accused Cambridge Analytica of specifically developing software tools, known as psychometrics, to target voters in the 2016 U.S. presidential campaign (National Public Radio, 2018). The use of these psychometrics to identify individuals or populations susceptible to socially engineered propaganda, in concert with the ability to manipulate neural areas of the brain to unconsciously shape behavior and disrupt decision-making abilities (i.e., neurocognitive hacking), foreshadows the sophisticated and potentially dangerous future of information operations in cyber conflict.

Neurocognitive hacking

The potential to secretly exploit adversary computer networks to prompt users with stimuli for the express purpose of manipulating neural structures and influencing political behavior is sobering. This capability is even more concerning as it can likely be accomplished using subliminal imagery of which a target audience is unaware. Because of its focus on manipulating actual neural structures in the brain, it goes beyond what has been conceptualized as “cognitive hacking,” a phenomenon more concerned with manipulating perception through the use of deception. For example, a common illustration of cognitive hacking involves the case of Mark Jakob, who created a series of false media releases to lower the cost of a specific stock and subsequently realized a significant profit (Cybenko et al., Reference Cybenko, Giani and Thompson2002). Although cognitive hacking can be covert and include the manipulation of perceptions, the concept does not address shaping behavior through the targeted activation of neural structures.

A simple way to conceptualize neurocognitive hacking is through the lens of a Russian propaganda sample used prior to the 2016 U.S. presidential election. Below is a highly inflammatory anti–Hillary Clinton advertisement that surfaced in the months leading up to the election. Although there is no way to quantify the amount of influence this or other ads may have had on the undecided electorate, it is possible that enough voters were swayed in critical states to have influenced the outcome of the election. Consider if, prior to viewing such an ad (Figure 2), subliminal, mortality-related images (Figure 1) were presented to activate the amygdalae and its associated neural components (i.e., threat response activation) of potential voters. Based on knowledge of the effects of mortality-related stimuli on in-group/out-group behavior, it is possible the images would have generated even greater emotion resonance and anti-Clinton sentiment on behalf of undecided voters.

Figure 1.

A dead body on a gurney is an example of a mortality-related stimulus. Shutterstock.com, royalty-free photo.

Figure 2.

An example of a Russian-produced propaganda meme found on Facebook prior to the 2016 U.S. Presidential election (Singer & Brooking, Reference Singer and Emerson2018).

PSYOPs personnel have been known to utilize disinformation campaigns (e.g., spreading rumors) to exacerbate underlying levels of societal conflict. And, with the advent of social media platforms, disinformation can spread more quickly and even generate lethal violence against perceived out-groups. For example, in early 2018, at least nine people were killed in India as a result of a rumor being spread in rural communities regarding the existence of a child abduction ring (Dwoskin & Gowen, Reference Dwoskin and Gowen2018). Using WhatsApp, tens of thousands of unsuspecting citizens spread video images of a faked child abduction, resulting in enraged local mobs attacking and killing innocent strangers suspected of involvement. Therefore, it is likely that the use of neurocognitive hacking techniques to activate threat-related neural structures of a targeted audience, prior to initiating a disinformation/rumor campaign, would increase both effectiveness and the rate of dissemination.

Additionally, strategically targeting specific members of a community who are more likely to believe conspiracy theories could increase both the rate and success at which the disinformation is spread. For instance, research suggests that political conservatives and less educated individuals are more likely to believe in conspiracies (Douglas et al., Reference Douglas, Sutton, Callan, Dawtry and Harvey2015; Fessler et al., Reference Fessler, Pisor and Holbrook2017). Furthermore, there is research suggesting that individuals who are 65 or older are seven times more likely to share disinformation on social media than younger individuals (Guess et al., Reference Guess, Nagler and Tucker2019). Thus, initially targeting older, less educated, and conservative members of a community with neurocognitive hacking attacks, followed by socially engineered, culturally appropriate disinformation, may increase the probability that the disinformation is ingested and disseminated throughout the community.

In remote tactical operations, in which military personnel need access to adversaries’ computer systems to disrupt command and control capabilities, neurocognitive hacking techniques could also be operationally useful. For example, subliminally prompting enemy forces with mortality-related stimuli may increase their propensity to open socially engineered emails containing malicious payloads, especially if the subject lines contain culturally resonant references.

In addition to utilizing subliminal mortality-related imagery, the use of subliminal sounds to induce or compound an effect is also a theoretical possibility. Research suggests that aversive sounds at specific acoustic frequencies can activate the human amygdala (Kumar et al., Reference Kumar, Kriegstein, Friston and Griffiths2012). Although this research did not investigate the effects of sounds in subliminal frequencies, exploring any possible compounding effects on behavior when combining subliminal visual and auditory prompts would be of interest, as the observed effects from visual prompts tend to last longer when given multiple times (Levy et al., Reference Levy, Pilver, Chung and Slade2014). Examples of effective mortality-related auditory prompts may include hissing sounds from venomous snakes or linguistic threats such as the words “death” and “kill.”

Future research

In light of revelations regarding Russian interference into U.S. and European Union elections, the ultimate aim of this line of inquiry is to better understand the extent to which behavior can be manipulated in response to subliminal prompts of mortality-related stimuli (M-rS) and explore means by which the effects can be mitigated (i.e., neurocognitive cybersecurity). These opportunities include investigating the effects of M-rS on hostile attribution, voting behavior, and group polarization. Given prior evidence suggesting M-rS produces effects similar to state anxiety (Gauthier, Reference Gauthier2011) along with evidence suggesting state anxiety can be compounded (Pederson & Larson, Reference Pedersen and Larson2016), exploring the relationship between priming frequency and behavior would be of interest (e.g., would prompting with 20 subliminal M-rS prompts every minute be more effective in eliciting targeted behavioral outputs than one every 10 minutes?).

Also, given that state anxiety is known to disrupt cognitive functioning (Eysenk et al., Reference Eysenk, Derakshan, Sanos and Calvo2007), exploring for decrements in reaction time, attention, and memory resulting from M-rS would be of interest. If negative effects were found, and given susceptibility to email phishing has been attributed to limitations in cognitive ability (Goel et al., Reference Goel, Williams and Dincelli2017), it follows that neurocognitive attacks could disrupt the speed and efficiency with which adversary military coders are able to efficiently respond to malware attacks during time-critical military operations.

Additionally, given previous research suggesting a correlation between the small ballistic eye movements referred to as micro-saccades and threat salience (Laretzaki et al., Reference Laretzaki, Plainis, Vrettos, Chrisoulakis, Pallikaris and Bitsios2011), along with the strong association between the human visual system and the amygdala (Burra et al., Reference Burra, Hervais-Adelman, Kerzel, Tamietto, Gelder and Pegna2013), further exploration of this connection would be of interest. A strong correlation may allow for the development of neurocognitive cybersecurity protocols for cyber operations personnel (i.e., detection of subliminal attacks using M-rS) as well as a biometric for use in medical diagnostics (e.g., Parkinson’s or Alzheimer’s disease) or security operations (e.g., detecting individuals with lethal ruminations/intent before boarding airplanes or entering military installations or entertainment venues).

An additional research avenue involves exploring the effects of repeated subliminal activations on the amygdala and ACC. In other words, much like the hippocampus of London taxi drivers has been shown to change after years of memory training (Maguire et al., Reference Maguire, Woollett and Spiers2006), it is logical to assume the amygdala and other neural components are similarly malleable. If so, determining whether amygdala growth moderates negative emotional reactivity (e.g., hostile attribution and ego threat) would be of interest.

Similarly, there is much research indicating the existence of neural correlates to common emotional experiences such as empathy, beauty, and romantic love (Lane & Nadel, Reference Lane and Nadel2002). Therefore, determining any repeated exposure effects to subliminal images known to activate these areas would be of interest and could extend the scope of neurocognitive hacking. For instance, would pairing subliminal images associated with positive emotional states with images of a cultural out-group help generate positive affect toward them? If so, how many subliminal exposures would be required to manipulate the specific neural structures and produce an effect? Hundreds an hour? Thousands a day? Millions over a period of weeks or months?

Finally, as evidence suggests that “mere exposure” to subliminal images of neutral objects can subsequently increase positive affiliation for them (Zajonc, Reference Zajonc2001), extending this line of research to neurocognitive hacking techniques may prove useful in civil-military operations. For example, consider a situation whereby the United Nations deployed a peacekeeping force composed of soldiers ethnicity different from the combatants they were ordered to separate. Theoretically, exposing the combatants to subliminal images of the ethnically different peacekeeping troops before the deployment may decrease the amount of suspicion or resistance the indigenous populations initially generate.