Concealed measures and manipulations of humility

When individuals answer questions about humility, they sometimes distort their responses, often to appear more appealing, or they cannot judge the degree to which they demonstrate this characteristic accurately.  The implicit association test is partly designed to overcome these concerns.

Illustration

Researchers often utilise the implicit association test (Greenwald et al., 1998), sometimes called the IAT, to measure humility or other traits or states, while the participants are oblivious to the purpose of this procedure. To illustrate this procedure, participants may first be instructed to press one button, perhaps the letter Q on a keyboard, whenever a word that corresponds to themselves appear, such as the words I, me, mine, or self.  They are also instructed to press another button, such as the P on a keyboard, whenever a word that corresponds to other people, such as they, them, their, it, or other, is presented.

Next, participants might be asked to also press the letter Q when a word that is synonymous with humility appears, such as humble, modest, tolerant, and respectful, and to press the letter P when a word that is synonymous with arrogance appears, such as immodest or egotistical (Rowatt et al., 2006).

Thus, in these trials, the same button represents both the self and humility.  Individuals who are humble—and thus associate themselves with humility—should perform this task effectively, committing few errors and responding rapidly. 

On other trials, the buttons that correspond to these traits are reversed. The letter P might correspond to words that epitomise the self and arrogance.  The letter Q might correspond to words that epitomise other people and humility.  In these trials, the same button represents both the self and arrogance. Individuals who are arrogant should thus perform more effectively in this condition. The difference in performance across the two conditions is assumed to reflect the extent to which individuals are humble (Rowatt et al., 2006).

Evidence the implicit association test can measure humility

Rowatt et al. (2006) assessed whether the implicit association test can indeed gauge humility accurately.  The results were encouraging.  For example

• the measure was fairly stable over time; the correlation between humility measured at one time and humility measured two weeks later was about .45,
• this measure of humility was inversely associated with several facets of narcissism, such as vanity and entitlement,
• this measure of humility was positively associated with the grade that students received in their course.

LaBouff et al. (2012) also utilised this procedure to gauge implicit humility.   Implicit humility predicted the likelihood that individuals would assist a person in need, even after controlling an explicit measure of humility. 

Similarly, Van Rensburg et al. (2022) applied the implicit association test to measure the personality trait that is often called honesty and humility.  However, this measure did not predict academic performance after controlling an explicit measure of this trait.  

Evidence the implicit association test can measure other characteristics

Few studies have explored whether the implicit association test can indeed gauge humility effectively.  Therefore, to evaluate the implicit association test, an examination of whether this procedure is generally accurate and effective may be useful.  Fortunately, a plethora of studies have explored, and usually validated, the utility of this procedure. For example, the test has been used to predict a variety of behaviours, such as acts of discrimination (McConnell & Leibold, 2001). A modified scoring algorithm that Greenwald et al. (2003) developed may augment the validity of this procedure.

Generally, correlations between implicit measures and explicit measures of some trait are small to moderate. Correlations between implicit measures of the five broad personality traits—extraversion, neuroticism, agreeableness, conscientiousness, and openness—and their explicit counterparts, as derived from the NEO-FFI ranged from .18 to .41 (Grum & von Collani, 2007). Other studies have also uncovered moderate correlations (e.g., Steffens & Buchner, 2003).

More importantly, several studies imply that implicit association tests seem to be relatively impervious to distortion. Asendorpf et al. (2002) showed that implicit association tests identified participants who were shy, even when these individuals were instructed to conceal this trait. The implicit association test can also identify prejudices even when participants were instructed to conceal these attitudes (e.g., Kim, 2003). However, Steffens (2004), at the University of Kaiserslautern-Landau in Germany, showed that faking can distort responses to some extent, but implicit tests are less susceptible to this problem than are explicit tests, even after practice.

Controversies

One criticism of the implicit association test is that individuals might orient their attention only to the desirability of items.  To illustrate, if individuals perceive themselves favourably, called a high implicit self-esteem (Bosson, Swann, & Pennebaker, 2000), they will tend to regard words such as I, me, mine, and self as desirable too.  Similarly, if individuals perceive humility as favourable, they will tend to regard synonyms of this term as desirable as well.  Consequently, when the same button corresponds to synonyms of the self and synonyms of humility, these individuals will perform well—even if they are not humble themselves. 

According to this argument, proposed by Steffans (2024), this implicit association test might partly reflect implicit self-esteem as well as reflect the extent to which individuals perceive the characteristic—in this instance, humility—as favourable.  Indeed, as Rowatt et al. (2006) showed, implicit humility was positively associated with implicit self-esteem.   However, implicit humility was positively associated with grades even after controlling implicit self-esteem.   Thus, implicit humility confers some insight beyond implicit self-esteem (Rowatt et al., 2006).

Challenges and the GNAT

Although useful, as McElroy-Heltzel et al. (2019) underscored, the implicit association test is more challenging to implement than are many alternative measures.  Participants need to use specialised software and maintain concentration over an extended duration. 

One variant of the implicit association test, called the Go No go association task or GNAT, is more efficient. Specifically, when participants complete the implicit association test, they need to press one of two buttons.  On each block of trials, one button might correspond to two concepts, such as self and humility, whereas another button might correspond to two other concepts, such as other and arrogance.  In contrast, when participants complete the GNAT, they press only one button whenever the targets appear, such as self and humility, and do not press this button when other words appear (e.g., Williams and Kaufmann, 2012). 

For each block of trials, the researcher can calculate d prime and B or leniency.  The measure of d prime, in essence, is the difference between the probability of correctly detecting the targets, called hits, minus the probability of incorrectly pressing the button when the words were not targets, called false alarms.  This difference is then subjected to an adjustment, called a probit.

A high d prime indicates the two concepts are appreciably associated with each other.  For example, suppose the d prime that corresponds to humility and negative as well as the d prime that corresponds to arrogance and positive is higher than d primes associated with humility and positive or arrogance and negative.  This pattern would indicate the individual perceives humility negatively but arrogance positively. 

The GNAT has been used to assess implicit associations in many circumstances.  In particular, this technique has been utilised to gauge the implicit attitudes towards spiders (Teachman, 2007), genetically modified food (Spence & Townsend, 2007), and racial prejudice (Smith et al., 2008).  The GNAT has also been utilised to gauge personality (Boldero et al., 2007).  However, the GNAT has not yet been used to measure humility.

Introduction

To some extent, the degree to which individuals demonstrate intellectual humility can be derived from the content of their conversations, such as how they express themselves during conflicts or debates.  For example, Hanel et al. (2023) proposed a framework that can fulfill this goal.  According to these researchers, individuals who exhibit intellectual humility demonstrate three inclinations during conversations and, especially, debates.  First, these individuals engage actively in the conversation, called qualified engagement.  For example, humble individuals may

• displays signs of support or camaraderie, such as nod their head or express a change in their state, such as “oh really”, called affiliation—while also agreeing or disagreeing with the other person,
• express agreement, such as expressing phrases like “absolutely”.

Second, individuals who exhibit intellectual humility seldom inflate the degree to which they feel certain—a certainty that is sometimes called boosted conviction.  For instance, humble individuals do not often

• express words that imply certainty, such as “definitely”, or “absolutely not”,
• express words that imply an event is ubiquitous, such as “always good” or “never bad”,
• imply that an assumption is obvious, epitomised by phrases like “of course”,
• imply that an assumption is true without evidence, such as “that fact is…”,
• imply that nobody could doubt an assumption, such as “how can you say that…”, or
• imply that an opposing position is unreasonable or ridiculous,

Finally, although humble individuals may express support and camaraderie, they do not display this affiliation unconditionally.  That is, they will express agreement and, to some extent, disagreement as well, with phrases like “that’s true but”.  In contrast, individuals who display affiliation but neither agreement nor disagreement are potentially behaving subserviently rather than humbly. 

The formula

The researchers then applied this framework to design a protocol or set of codes to measure the intellectual humility of debaters.  To generate these codes, a panel of linguists analysed 20 recorded debates.  Specifically, the panel defined the following set of five codes:

• affiliate-agree: gestures, comments, or other verbal cues that indicate both camaraderie and agreement, such as “yeah-exactly”,
• affiliate-disagree: gestures, comments, or other verbal cues that manifest collaborative disagreement, such as “yeah but” or “that’s true but”,
• agreement: gestures, comments, or other verbal cues that imply agreement but without any hint of affiliation, support, or camaraderie, such as “I agree”,
• affiliation:  gestures, comments, or other verbal cues that demonstrate affiliation, support, or camaraderie but without any hint of agreement or disagreement,
• boosted conviction, such as gestures, comments, or other verbal cues that imply certainty.

To measure intellectual humility, the researchers then utilised a formula.  Specifically

• the numerator is affiliate-agree + affiliate-disagree + agreement – boosted conviction – affiliation,
• the denominator is 1 + boosted conviction + affiliate-agree + affiliate-disagree + agreement.

In essence, the numerator represents the degree to which the person agrees, or disagrees collaboratively, but not display unconditional certainty or agreement.  The denominator represents the number of interactions in which individuals could have displayed these tendencies. 

Evidence that substantiates this measure

Hanel et al. (2023) also collated some evidence that corroborates the validity of this measure.  First, in one study, comprising over 300 participants, about half these individuals completed a list of 19 values—such as development, community, tradition, or power—and indicated which of the five values they perceive as most important.  Then, these individuals wrote about why one of these values was important and how this value guides their behaviour.  This exercise is often called self-affirmation.  In the control condition, participants wrote about why they like or dislike five beverages.  

The authors hypothesised that self-affirmation should foster humility.  That is, after individuals complete this exercise, they should feel more certain of their identity, diminishing a sense of fragility.  As this fragility dissipates, these individuals may not be as vulnerable to information that may challenge this perception of themselves.  They should, therefore, be more willing to acknowledge their limitations or misconceptions, manifesting as humility.  Consistent with this possibility, as Hanel et al. (2023) revealed, if participants had completed the exercise that promotes self-affirmation, they exhibited greater humility during subsequent debates—as measured by the previous formula. 

Introduction

Rather than invite participants to answer questions or to complete a range of computer tasks, another method that researchers or practitioners could utilise to assess humility is to observe the behaviour of individuals.  For example, according to Van Tongeren et al. (2023b), researchers or assessors could first arrange a setting in which some people exhibit humility and other people exhibit arrogance or subservience—such as activities in which

• individuals need to negotiate on some matter,
• individuals deliver feedback to each other after completing some task,
• individuals act as a leader who needs to instruct someone to work after hours,
• individuals need to interact with someone from another ethnicity or field,
• teams need to solve a problem and each member is assigned a unique role,
• teams need to discuss a contentious matter and nobody is officially the leader.

During these activities, to uncover signs of humility, researchers or assessors could ascertain, for example, which individuals

• support the ideas that other members suggest,
• assign themselves a role that is neither superior nor subservient to anyone else,
• present the strengths and limitations of all cultures or fields, including their own.

Van Tongeren et al. (2023) also enumerated other manifestations of intellectual humility, relational humility, or cultural humility that researchers could assess.  Some of these behaviours revolve around interpersonal behaviours, such as instances in which individuals

• embrace information that diverges from their beliefs or atttiudes rather than behave defensively,
• praise other individuals,
• accept responsibility for errors,
• demonstrate respect towards diverse people or perspectives,
• show empathy and understanding,
• seem eager to learn from other people.

Other behaviours may reveal how individuals evaluate themselves , such as instances in which participants acknowledge strengths, limitations, and biases, recognise their beliefs may be incorrect, or concede when they are uncertain about some topic.

To be effective, these behavioural measures should fulfill at least four criteria (Van Tongeren et al. (2023b):

• First, researchers should be able to administer these assessments to many individuals as efficiently as possible—perhaps using virtual interactions (cf Richard & Lauterbach, 2004) and machine learning to evaluate performance (e.g., Hickman et al., 2022).
• Second, researchers would need to identify features of humility that are observed across cultures or circumstances as well as features of humility that are specific to particular cultures or circumstances.
• Third, because humility is not as observable as some attributes, such as aggression, researchers would need to identify patterns of responses, such as particular micro-expressions (cf Foley & Gentile, 2010; Wu et al., 2011), that are sensitive to humility.
• Fourth, researchers need to uncover patterns of responses that are specific to humility—rather than inadvertently assess other similar attributes, such as agreeableness. 

Benefits and drawbacks

Van Tongeren et al. (2023a) outlined some benefits of these behavioural assessments.  First, to thrive on these tasks, individuals need to adapt their behaviour to the circumstances.  Rather than merely adhere to a rehearsed script and, for example, merely acknowledge their limitations, participants would need to demonstrate humility at an appropriate time, with an appropriate strategies.  They would not, for example, acknowledge their faults before they have established credibility with a conceited staff member.  Second, these assessments preclude the tendency of people to inflate their humility or to be oblivious to their level of humility.   

Nevertheless, Van Tongeren et al. (2023a) also conceded some drawbacks to these behavioural assessments, besides the additional time and resources that researchers would need to mobilise.  First, to quantify the humility of participants, researchers would need to develop a coding scheme that is valid and also sensitive to the culture of these individuals.   Second, individuals often behave differently when their behaviour is observed.

Rather than measure humility, some researchers prime or manipulate the level of humility in participants.  This approach enables researchers to ascertain the direction of causality—that is, to determine whether humility generates some outcome rather than vice versa.

For example, to explore whether humble people tend to distribute resources fairly, Wang et al. (2025) introduced a procedure that primes humility.  First, participants read a story that was customised to resonate with Chinese participants.  The story epitomised a humble person.  Here is the story:

”…Zhang Ming was awarded the Principal’s Scholarship for his outstanding academic performance and was invited to give a brief speech to the graduating students. In his speech, he attributed his achievements to the help of his teachers and the support of his friends. When interviewed, Zhang Ming’s classmates all mentioned that he never boasted about himself and was always willing to help his classmates academically. When interviewed directly, Zhang Ming stated that he could recognise both his strengths and weaknesses and emphasized the importance of continuously improving his shortcomings to strive for better…”

Next, to prime humility, participants were prompted to summarise, in one word, the quality that Zhang Ming epitomises.  Next, participants were asked to indicate, on a scale from strongly disagree to strongly agree, the extent to which Zhang Ming is humble and the degree to which humility is an important virtue.  Finally, participants reported a personal experience of humility.  Participants in the control group reported their daily activities at school.  If humility was primed, participants were more inclined to distribute resources fairly between themselves and other people.