Influence of Category Label and Metaphor on Judgments About Mental Disorder Characteristics

Category labels affect people’s judgments regarding mental disorders which are unknown to them. Descriptions of these ‘unknown’ disorders that do have a name, are assumed by people to be more stable and having reasons to exist, when compared with the same descriptions of disorders - without a specific name [3]. However, it is not clear whether this effect can be evoked by other linguistic parameters, for instance, by metaphors. We hypothesized that including a metaphor in the description of a mental disorder would lead to the same effect even without a category name. We replicated a study by Giffin and colleagues’ and added a new experimental condition in which participants read texts with the descriptions of a person’s unusual behaviour without the disorder's name, but with its metaphoric description. After reading the texts, participants assessed a few statements concerning some characteristics of the disorder. The results showed that the effect of a category label was replicated, and the metaphoric description also evoked a significant effect, but it was found in judgments of different characteristics of the disorder.

Introduction

When people need to evaluate a phenomenon and make a judgment on it, they rely on the additional information available to them at the time they evaluate the phenomenon. Research shows that people find explanations of psychological phenomena more meaningful and plausible if they are based on additional scientific information, even if that information is irrelevant to the explanation [13], and they tend to evaluate judgments as more true if they contain scientific terminology, even though to them, it does not make any sense [10]. These findings suggest that people have cognitive biases formed by various factors, particularly, the use of special names.

Indeed, the presence or absence of a special name affects how people judge phenomena and influence which features are perceived first [7]. For example, in a study of Giffin et al. [3], participants were shown descriptions of culturally specific mental disorders. The authors constructed four vignettes, each describing a person with unusual behaviour. The participants read one of the four texts and then evaluated several statements about the behaviour described in the text. One group read the texts in which the unusual behaviour was described with a fictional category label ("depathapy"), while another group read the same texts, but without a label. Participants who read the texts with the label, were less likely to blame the person with the disorder for their unusual behaviour, and were more likely to believe that both they and others would behave the same way if they had this disorder, and were more likely to believe that the disorder has biological rather than psychological causes.

Giffin et al. showed that an additional label for the phenomenon changes its evaluation [3]. However, they did not describe whether this effect can be caused by alternative linguistic means. The purpose of the current paper is to clarify the linguistic parameters of this effect. It is known that the general designation for a group of phenomena can be presented to a person in the form of special categorical names (for example, "modular psyche"), or with the help of metaphors (for example, psyche as "Swiss knife"). A metaphor, as a figurative comparison, is often used in the explanation of phenomena, especially if these phenomena are complex, not obvious, and helps to explain
a new concept through already formed knowledge [5; 9; 11]. Some studies show that metaphors can also influence judgments about different diseases. Specifically, the metaphor of travel increases the willingness to accept the difficulties of treatment, which is not the case with the metaphor of battle [4]. Another study shows that a description of
a state using a body metaphor (e.g., likening the heart to the capital and arteries to roads), influences people's judgments about a virus spreading in a country [6]. This metaphor strengthens the tendency for subjects to agree that a virus is dangerous for their health, that special hygiene procedures must be carried out to prevent and control the virus, and that special policy measures must be taken to fight the virus.

Some studies have noted the importance of metaphorical information in communication concerning mental disorders. For example, it has been argued that metaphors can promote understanding between psychiatrists and patients when discussing the effects of drugs [1]. According to other studies, metaphors help some patients describe their psychotic experiences in a more comprehensible way, while at the same time they can be used as a tool for psychotherapy [8].

In order to compare the impact of a categorical label on judgments about psychological disorders with an effect that can potentially be triggered by a metaphor, we replicated the Giffin et al. experiment adding an additional experimental condition that did not have a name for the disease, but did have a metaphor [3]. We hypothesized that, as in Giffin’s study, the category label would lead participants to judge the labeled phenomenon (in our case – about behaviour) as more stable and generalizable, in comparison with an unlabeled phenomenon. We also proposed that a metaphor would affect people’s judgment in a similar way toward a category label.

Method

Participants. Two-hundred-and-twenty-four young adults (54% – female, mean age – 20.1 years, SD = 4.65 years) participated in the study. All participants were university students who were not studying psychology, were older than 17, lived in Russia, and were Russian native speakers. They received extra credit for an introductory course in psychology for their participation in the study.

Materials and procedure. Four texts (4-6 sentences each) with descriptions of unusual human behaviour were created, in Russian, adapted from the study by Giffin et al. [3]. The text followed two conditions: first, these categories should not be familiar to the participants; second, they should look and sound like other similar related terms and be constructed using natural language. For these reasons, the authors elaborated behavioural categories based on the symptoms of culturally specific disorders.

The last sentence in each text explained the unusual behaviour of the character in the text and differed between conditions. In the named condition, the described behaviour was given a name, whereas in the tendency condition the behaviour was described as tendency and did not have any category label. All four behavioural categories in the named condition were named in the texts using the artificial word, depathapy. We replaced the text characters’ names with common Russian names, and translated the word depathapy literally as депатафия.

As a result, in the replication we used the word депатафия (depathapy) in the experimental condition (category label condition), whereas in the control condition we used the word склонность (tendency) (e.g., It turns out that David has Depathapy,
a tendency to imitate the actions of others and obey commands directed at them, leading him to take the painting // It turns out that David has a tendency to imitate the actions of others and obey commands directed at them, leading him to take the painting).

We added another experimental condition in which a category label was absent but
a metaphorical description of the behaviour was included (как по чьей-то команде; as if on someone’s command; e.g., It turns out that David has a tendency, as if on someone’s command, to imitate the actions of others and obey commands directed at them, leading him to take the painting). The reason for using this metaphor is based on the idea from Giffin’s study, that the category label shifts responsibility for the action of the subject, to the phenomena, which is expressed by the category label [3].

One of the four texts with modifications for all conditions (category label, metaphor and control) is presented below as an example:

 “David is a 40-year-old male. Recently, he took a beautiful and expensive painting from his office after one of his coworkers said, ‘you should take that painting, you’re the only one who ever looks at it.’ David’s coworker had not been serious.

[Сategory label] It turns out that David has Depathapy, a tendency to imitate the actions of others and obey commands directed at them, leading him to take the painting.

[Metaphor] It turns out that David has a tendency, as if on someone’s command, to imitate the actions of others and obey commands directed at them, leading him to take the painting.”

[Control] It turns out that David has a tendency to imitate the actions of others and obey commands directed at them, leading him to take the painting.”

Each text was followed by 13 statements which participants assessed using a 7-point scale. These statements tested participants’ attitude to different characteristics of the described phenomena (i.e. depathapy/tendency) to reveal whether it was perceived as
a real disorder, or as a behaviour based on subjective causes. The authors of the original study did not find differences between conditions in all statements. They did not explain what caused such a result, which is why we used all the statements from the previous experiment, in the current one.

The experiment was held online using 1ka platform (www.1ka.si). Participants were randomly assigned to one of four texts. After reading one text, participants were presented with 13 statements divided into blocks, and were asked to assess them using the 7-point scale.

The first block included statements about the plausibility of the explanation of the character’s behaviour presented in the texts, whether he or she should be blamed and is guilty for the action taken.

As an example, the statements related to a text are presented below. The statements were the same for each condition, but in the category label condition, the name of the behaviour was included in the statements, whereas in the metaphor and control conditions the word tendency was used instead of the name of the behaviour (Table 1).

The next block contains statements that tested whether the disorder name could implicate such characteristics of behaviour as stability in the past and future, and also its generalizability to other people and to the participant (Table 2).

Table 1

Statements about plausibility of a given explanation, blame and guilt
of a character with unusual behaviour

Table 2

Statements about stability of the behaviour in time and its generalizability to others or to the participant

The third block included statements about the probability of the described behaviour to have a biological or psychological nature. These statements were followed by short texts explaining what is meant by biological and psychological factors (Table 3).

Table 3

Statements about biological and psychological status of the behaviour

The next block included statements assessing the probability of effectively treating the unusual behaviour using medication or psychotherapy. The statements were prefaced by a short paragraph describing the meaning of medication and psychotherapy (Table 4).

Table 4

Effectiveness of medication and psychotherapy to treat and control the behaviour

The last block of statements tested whether the presence of the disorder name leads to the tendency to think the described behaviour has a common cause or common symptoms (Table 5).

Experimental design. The experiment had a between-subject design to avoid mixing effects evoked by different conditions (i.e., by category label, metaphoric description or the absence of both). There were two experimental groups and one control group, independent variables were category label and metaphor (or their absence), and dependent variables included – participants’ assessments of each statement.

Table 5

Existence of common cause and symptoms for the behaviour

Results

A Kruskal–Wallis test showed the effect on judgments about explanation
(χ²(2) = 15.05, p < .001), blame (χ²(2) = 14.03, p < .001), legality (χ²(2) = 6.56, p = .038), generalization to others (χ²(2) = 15.32, p < .001), generalization to self (χ²(2) = 17.29,
p < .001), and medication (χ²(2) = 19.12, p < .001). It also showed a tendency towards
a difference in judgments about stability in the future (χ²(2) = 5.05, p = .08). For other differences the p-value was more or equal to .132. We used non-parametric statistics because the assumption of data for normal distribution was violated (Shapiro–Wilk test for each variable had p < .001).

We compared mean responses for the answers in three conditions by pairs, using
a post-hoc Dunn test with Holm adjustment. The label “depathapy” increased the objectivity of the category: compared with the control condition, the participants from the category label condition were less likely to blame the person for destructive actions
(p = .006), and they assumed that other people (p < .001), and they themselves (p < .001) would behave the same way in similar circumstances. They also believed that this condition is manageable with medical therapy (p = .004; see Table 6). These results replicate the effect of a category label found in Giffin et al’s study [3].

As we expected, the metaphor had a framing effect on the categories of mental disorders: it reduced the persuasiveness of the explanation by referring to a person’s state (p = .001), and increased the belief that this condition will continue in the future (p = .039), and that it can be treated with medication (p < .001).

When we compared answers in the two experimental conditions, we also found significant differences. In the category label condition (in comparison with the metaphor condition), the participants blamed the person less for any destructive behaviour (p < .001) and believed that he or she was less guilty in terms of the law (p = .016); they also assumed that other people (p = .006) and they themselves (p = .006) would behave in the same way if they had a similar disorder. Also, an explanation of a person’s behaviour by his or her mental condition was less satisfying in the case of the metaphor condition (p = .001).

Таble 6

Descriptive statistics of answers in different experimental
(category label and metaphor) and control groups

Notes. Bold marks the differences between the experimental (category label and metaphor) and control conditions. Italics marks the differences between the experimental conditions. Both bold and italics marks differences between both experimental and control conditions. * – p < .05, ** – p < .01, *** – p < .001.

Discussion

We investigated whether disorders that have a name are evaluated as more stable and having reasons to exist, compared with the same description of a condition, but without a specific name (replication of Giffin et al. study [3]). We also studied whether this effect can be evoked by other linguistic parameters, such as metaphors. First, we showed that the label of a culture specific disorder can influence judgments about this disorder, which partly replicates the results of Giffin et al. [3]. Second, we showed that metaphors can trigger the same effect and have an influence on judgments about disorders previously unknown to the participants. Use of the metaphorical feature, ‘as if on someone's command’, influenced categorization scores by enhancing the idea of objectivity of culturally specific mental disorders. This confirms the main hypothesis: the metaphor caused an effect that is partially identical to the effect of a categorical name (in the question of medication). The influence of the label and the metaphor varied in semantic value and was manifested in the assessment of different characteristics of the disorder. Interestingly, a metaphor that "shifts" responsibility for pathological malicious behaviour to an "external" object (or subject) strengthened the belief that the person's behaviour will not change over time, and that it can be changed through external unilateral intervention
(e.g., medication).

It was unexpected that the metaphor reduced the persuasiveness of explaining behaviour through an appeal to human propensity. On the one hand, it may be related to the semantics of the metaphor: if a person's behaviour is caused by "someone's command", it cannot be explained by the characteristics of this person. On the other hand, here we may face the limitations of the epistemological potential of unconventional metaphors, which, according to the theory of "metaphor career", functions as a low-level generalization based on comparison rather than categorization, and cannot completely replace the label in cognitive processes [2]. This reasoning provides direction for further research into the impact of metaphors on personal judgment.

We also need to mention several limitations of our study. First, we understand that we used artificial categories which could differ from natural categories. Second, we used information about culturally specific diseases, so the participant learned a new category during the experiment. But does language influence judgments about familiar mental disorders? And does the influence of the label and metaphor fade over time? Obviously, knowledge about how language affects the perception of mental disorders needs to be expanded in the future.

Note also that the results of this experiment show that there is a possibility of metaphorical framing when talking about categories of mental health conditions. The effects of using metaphors haven't been investigated previously in terms of how they affect people's judgment on those who have such conditions. For this reason, the findings of our study support the list of categories, for which it has been empirically proven, that metaphors may affect judgment about them [12].

From an applied perspective, it also means that metaphors used in communication about complex psychological conditions (e.g., disorders) not only highlight certain features, but may also have some influence on attitudes toward people with such conditions. Thus, in professions directly related to such communications (psychotherapists, psychiatrists or even journalists), it is necessary to be sensitive to the choice of means of language expression so that metaphors, while fulfilling their cognitive function (facilitating understanding), do not lead to stigmatization.

SUPPLEMENTARY MATERIALS

Text with a description of unusual behaviour

1.             Elena is a 40-year-old female. Recently, she screamed at and hit her boss when he approached her about a project she was working on. Her boss had to be taken to the hospital. Co-workers reported that after the incident, Elena was trembling. It turns out that Elena has Depathapy, a tendency to tremble and act verbally and physically aggressive, leading her to hit her boss // It turns out that Elena has a tendency to tremble and act verbally and physically aggressive, leading her to hit her boss// It turns out that Elena has a tendency, as if on someone’s command, to tremble and act verbally and physically aggressive, leading her to hit her boss

2.             Mark is a 40-year-old male. Recently, he broke into several of his neighbors’ houses, taking various items – from napkin holders to vases. When the police found him, he seemed to believe each object was highly valuable. It turns out that Mark has Depathapy,
a tendency to steal objects believing them to be of high value, even though they seldom are // It turns out that Mark has a tendency to steal objects believing them to be of high value, even though they seldom are // It turns out that Mark has a tendency, as if on someone’s command, to steal objects believing them to be of high value, even though they seldom are.

3.             Rimma is a 40-year-old female. Recently, she was seen in the street without a shirt or pants on. The police took her into custody. At the police station, she began breaking furniture and objects, and tried to run from the building. It turns out that Rimma has Depathapy, a tendency to remove clothing, break furniture, flee from shelter, and perform other irrational or dangerous acts // It turns out that Rimma has a tendency to remove clothing, break furniture, flee from shelter, and perform other irrational or dangerous acts // It turns out that Rimma has a tendency, as if on someone’s command, to remove clothing, break furniture, flee from shelter, and perform other irrational or dangerous acts.

1. Amarasingham L.R. “This will clear your mind”: The use of metaphors for medication in psychiatric settings. Culture, Medicine and Psychiatry, 1984. Vol. 8, no. 1, pp. 49–70.
2. Bowdie B.F., Gentner D. The career of metaphor. Psychological Review, 2005. Vol. 112, no. 1, pp. 193–216. DOI:10.1037/0033-295X.112.1.193
3. Giffin C., Wilkenfeld D., Lombrozo T. The explanatory effect of a label: Explanations with named categories are more satisfying. Cognition, 2017. Vol. 168, pp. 357–369. DOI: 10.1016/j.cognition.2017.07.011
4. Hendricks R.K., Boroditsky L. Emotional implications of metaphor: consequences of metaphor framing for mindset about hardship. Proceedings of the 38th Annual Conference of the Cognitive Science Society, 2016. pp. 1164–1169.
5. Landau M.J., Meier B.P., Keefer L.A. A metaphor-enriched social cognition. Psychological Bulletin, 2010. Vol. 136, no. 6, pp. 1045–1067. DOI: 10.1037/a0020970
6. Lu H., Schuldt J.P. Communicating Zika Risk: Using metaphor to increase perceived risk susceptibility. Risk Analysis, 2018. Vol. 38, no. 12, pp. 2525–2534. DOI: 10.1111/risa
7. Lupyan G. Linguistically modulated perception and cognition: The label-feedback hypothesis. Frontiers in Psychology, 2012. Vol. 3, no. 54. DOI: 10.3389/fpsyg.2012.00054
8. Mould T.J., Oades L.G., Crowe T.P. The use of metaphor for understanding and managing psychotic experiences: A systematic review. Journal of Mental Health, 2010.
   Vol. 19, no. 3, pp. 282–293. DOI:10.3109/09638231003728091
9. Murphy G.L. On metaphoric representation. Cognition, 1996. Vol. 60, no. 2,
   pp. 173–204. DOI: 10.1016/0010-0277(96)00711-1
10. Pennycook G., Cheyne J.A., Barr N. et al. On the reception and detection of pseudo-profound bullshit. Judgment and Decision making, 2015. Vol. 10, no. 6, pp. 549–563.
    URL: https://www.sas.upenn.edu/~baron/journal/15/15923a/jdm15923a.pdf (Accessed: 12.08.2020).
11. Speed L.J., O’Meara C., San Roque L. et al. Perception Metaphors. n.d. № 17.
    URL: https://benjamins.com/catalog/celcr.19 (Accessed: 12.02.2019).
12. Thibodeau P.H., Hendricks R.K., Boroditsky L. How linguistic metaphor scaffolds reasoning. Trends in Cognitive Sciences, 2017. Vol. 21, no. 11, pp. 852–863. DOI: 10.1016/j.tics.2017.07.001
13. Weisberg D.S., Hopkins E.J., Taylor J.C. People’s explanatory preferences for scientific phenomena. Cognitive Research: Principles and Implications, 2018. Vol. 44, no. 3. URL: https://link.springer.com/content/pdf/10.1186/s41235-018-0135-2.pdf (Accessed: 12.08.2020).