Matthew S. McGlone

Forewarning and Forearming Stereotype-Threatened Students

This study investigated communicative strategies for helping female students cope with “stereotype threat”. Participants completed a difficult math test after reading one of three coping messages: a control message encouraging perseverance, a “suppression” message describing stereotype threat and instructing participants to suppress associated thoughts, and a “replacement” message describing the phenomenon and presenting an alternative, self-relevant positive stereotype. As predicted, a performance gap favoring males was observed in the control condition. This gap widened when students attempted to suppress threatening thoughts but narrowed when they were primed with the alternative positive stereotype. Our results demonstrate that priming a positive achieved identity (e.g., private college student) can subdue stereotype threat associated with an ascribed identity (e.g., female).

Hyperbole, Homunculi, and Hindsight Bias: An Alternative Evaluation of Conceptual Metaphor Theory.

To its credit, conceptual metaphor theory (CMT) has drawn significant attention to the question of what figurative language can tell us about human concepts. However, the answers CMT theorists have offered are typically unsubstantiated by the empirical evidence, and occasionally unfalsifiable. This reply to Raymond W. Gibbs Jr.s positive evaluation of the theory offers an alternative assessment that is more critical of its shortcomings. This critique summarizes four basic problems with CMT: The theory (a) is attributionally ambiguous about the locus of metaphoric motivation, (b) commits a form of the infamous “homunculus problem” in philosophy of mind, (c) employs circular reasoning to formulate hypotheses and interpret linguistic evidence, and (d) is not parsimonious. All of these problems are evident in Gibbss piece, and thereby undermine his defense of CMTs explanatory value.

Vicious Viruses and Vigilant Vaccines: Effects of Linguistic Agency Assignment in Health Policy Advocacy

Studying the effect of a fictitious policy editorial advocating mandatory vaccination of youth against human papillomavirus (HPV), the authors hypothesized that linguistic assignment of agency to HPV (e.g., “HPV preys on millions of people”) would increase perceptions of its severity, relative to a comparable message that assigned agency to humans (e.g., “Millions of people contract HPV”). In addition, the authors predicted that HPV vaccines would be perceived as more effective when agency was assigned to vaccination (e.g., “Vaccination guards people”) rather than to humans (e.g., “People guard themselves through vaccination”). University students (N = 361) were randomly assigned to read one of four versions of the editorial defined by a 2 ×2 (Threat Agency × Immunization Agency) factorial design and thereafter completed a questionnaire. When agency was assigned to the virus or the vaccine, HPV was perceived as a more severe threat, vaccination was perceived as more effective, and people were more in favor of mandatory HPV vaccination. The authors concluded that linguistic agency assignment bestows potency to the agent, thereby making threats more alarming and medical interventions seem more effective.

The Continuum of Metaphor Processing

We describe the explanatory value of a relativistic account of metaphor processing in which different modes of metaphor interpretation are assumed to be operative in different discourse contexts. Employing the cognitive psychological notion of a processing set, we explain why people might favor attributional interpretations of figurative expressions in some circumstances and analogical interpretations in others. Applying this logic to findings in the psycholinguistic literature on metaphor suggests that some of the competing models may in fact describe different points on a continuum of metaphor processing.

Persuasive Effects of Linguistic Agency Assignments and Point of View in Narrative Health Messages About Colon Cancer

The authors explored the effects of linguistic agency and point of view on narrative force. Participants (N = 499) were randomly assigned to read one version of an article about colon cancer, defined by a 2 (disease agency: cancer, human) × 2 (temporal agency: death, human) × 2 (point of view: first person, third person) between-subjects design. Disease agency language assigned agency to cancer (e.g., “Cancer developed in me”) or to humans (e.g., “I developed cancer”). Temporal agency language described death as approaching humans (e.g., “as death closes in on patients) or as being approached by humans (e.g., “as patients close in on death”). The narrative was presented from the first-person singular or third-person plural viewpoint. Participants then completed a questionnaire measuring threat perceptions, efficacy, transportation, and other study variables. Language assigning agency to humans rather than to cancer elevated susceptibility beliefs. Death-approach language led to greater fear than human-approach language without impacting efficacy perceptions. Human-approach language was rated more persuasive than death-approach language, but only in first-person point-of-view narratives. Transportation and identification were positively associated with ratings of threat severity and susceptibility, fear, efficacy, behavioral intentions, and message persuasiveness. Implications for message design are discussed.

Giving Radon Gas Life Through Language: Effects of Linguistic Agency Assignment in Health Messages About Inanimate Threats

Health messages that linguistically assign agency to a threat (e.g., HIV infects people) tend to evoke more fear and elevate perceptions of threat severity and susceptibility relative to those that assign agency to humans (e.g., people contract HIV). The present experiment (N = 843) extended these findings to a nonliving health threat, radon gas, and compared nonsentient (e.g., radon gas is seeping . . .) and sentient (e.g., radon gas is invading . . .) threat agency language. Sentient threat agency language elevated perceptions of threat severity compared to both nonsentient threat and human agency language, which did not differ from each other. Furthermore, sentient threat agency language evoked more fear than nonsentient agency language when the advocated recommendations were moderately (but not completely) effective.

Laying It on Thin: Analogical Cue Frequency in the Manipulation of Choice

Persuasive analogies consist of linguistic cues that invite audiences to represent a problem in terms of an analog and to make choices compatible with this manipulated representation. The authors explore how the frequency of linguistic cues moderates analogical bias in choice behavior. Participants read versions of a managerial decision scenario differing in the number of sports (e.g., game plan) or family (e.g., parent company) idioms used to describe situational attributes and then chose between analogically consistent and inconsistent response options. Analogies consisting of an intermediate number of cues elicited higher rates of analogically consistent responding than those of higher and lower cue frequency. When explaining their responses, participants referred to the sports or family analogies most often to rationalize rejecting the analogically consistent response option after reading high cue frequency versions. These findings indicate that there are lower and upper boundaries to a persuasive analogys optimal cue frequency.

Targeting Type 2: Linguistic Agency Assignment in Diabetes Prevention Policy Messaging.

We explored the effects of linguistic agency assignment on the persuasive impact of a fictitious medical journal editorial about Type 2 diabetes. Participants (N = 422) read 1 of 4 versions of an editorial that differed in the language used to describe the health threat posed by the disease (threat agency) and to outline a program for preventing it (prevention agency). Threat agency was assigned either to the disease (e.g., diabetes puts individuals’ lives at risk) or to humans (e.g., individuals who acquire diabetes put their lives at risk). Prevention agency was assigned either to the recommended prevention behaviors (e.g., a healthy diet and regular exercise protect children from Type 2) or to humans (e.g., children who eat a healthy diet and exercise regularly protect themselves from Type 2). Respondents’ perceptions of disease severity were higher when threat agency was assigned to diabetes rather than humans. However, attitudes toward the proposed prevention program were higher when prevention agency was assigned to humans rather than to the recommended behaviors. The latter finding contrasts with agency effects observed in previous research on a viral threat, suggesting that the optimal pattern of agency assignment in prevention messaging may be different for acute and chronic lifestyle diseases.

Stereotype Threat and the Evaluative Context of Communication

We explored the impact of “stereotype threat” —that is, distress associated with the prospect of confirming a negative stereotype—on communication in evaluative contexts. Participants engaged in a conflict resolution simulation framed as diagnostic of their ability either to be a leader or to maintain close personal relationships. Women were less fluent and used more tentative language under leadership than relational maintenance framing, but men were less fluent and more tentative under relational maintenance than leadership framing. The influence of stereotype frame on the rates of disfluencies and tentative language was partially mediated by state anxiety. Our findings demonstrate that the effects of situationally induced stereotype threat on communication behavior are comparable to its effects on intellectual test performance. Consequences of stereotype threat for impression formation and strategies for reducing its impact on social interaction are discussed.

Adversarial Infrahumanization in the Abortion Debate

Primary emotions such as pleasure or fear are routinely attributed to both humans and animals, but secondary emotions such as pride or resentment are typically attributed only to humans. Infrahumanization occurs when secondary emotions are attributed to human ingroups but not outgroups, thereby portraying outgroup members as less than human. This phenomenon is well documented as a psychological process, but has not been investigated in communication. The present study explored communicative infrahumanization in the abortion debate. Analyses of the emotion language used by prolife and prochoice bloggers revealed a reliable tendency among both groups to attribute primary but not secondary emotions to their adversaries.