Tipultech logo

The empathy gap effect

Author: Dr Simon Moss


Individuals often underestimate the power of urges and feelings--such as pain (Nordgren, van der Pligt, & van Harreveld, 2006), hunger (Nordgren, van der Pligt, & van Harreveld, 2007), sexual arousal (Ariely & Loewenstain, 2006), fatigue (Nordgren, van Harreveld, & van der Pligt, 2009), and cravings (Sayette, Loewenstein, Griffin, & Black, 2008)--on their behavior. They might, for example, predict they will be readily able to curb their urge to smoke in a social setting, like a pub. When they arrive in this social setting, however, the urge is often stronger than perhaps they predicted.

This phenomenon is called the cold-to hot empathy gap (Loewenstain, 1996) or simple the empathy-gap effect (Nordgren, van Harreveld, & van der Pligt, 2009). According to Loewenstein (1996), when an urge is dormant rather than activated, individuals cannot readily evoke the visceral experience of this urge. When some ache has dissipated, for example, they cannot invoke the feeling of pain they had earlier experienced. When their hunger has abated, they cannot readily elicit the visceral experiences--the mild throbbing around the stomach region or tightness around the chest--this hunger may have evoked.

Because individuals cannot readily invoke these visceral experiences, they underestimate the intensity of these urges. They feel these urges can be readily suppressed or retrained. In short, in a cold state--a state in which they urge or feeling has dissipated--individuals cannot evoke the visceral or bodily experiences they endured in a hot state--a state in which the urge or feeling had been evoked.

Empirical evidence

Effect of physical pain

One of the most compelling illustrations of this empathy gap effect was reported by Nordgren, van Harreveld, and van der Pligt (2009). In their study, participants completed an intellectual test that assessed their memory. Participants were instructed to insert their hand in icy water. The temperature of this water tends to impair performance on this task.

Later, participants were asked to specify the extent to which various factors, including the icy water, had stifled their performance. While they specified these factors, only some, but not all, of the participants were instructed to insert their hand in icy water. If their hand was currently inserted in icy water, they maintained the temperature greatly impeded their performance on the memory task. If their hand was not inserted in icy water, they did not feel the temperature appreciably affected their performance on the memory task.

Presumably, when their hand was removed from the icy water, they could not evoke the visceral reactions to this act. They could not elicit the numbness in their fingers or the discomfort across their wrist, for example. Hence, they underestimated the impact and intensity of these bodily experiences.

Effect of social pain

As Nordgren, Banas, and MacDonald (2010) showed, because of the empathy gap, individuals tend to underestimate the pain that other people feel when they are ostracized or excluded. That is, ostracism and exclusion can be remarkably painful. When individuals experience this pain, they appreciate the suffering of other people who also feel ostracized, excluded, shunned, or disdained. However, when they do not experience this pain, they underestimate the distress that exclusion can evoke.

Nordgren, Banas, and MacDonald (2010) demonstrated this bias. In a series of studies, some participants were either excluded or not excluded in a social setting. Next, they rated the degree to which other people are likely to feel negative emotions when excluded. Relative to participants who were excluded, participants who were not excluded underestimated the negative emotions that other people would feel when shunned. Indeed, in one study, participants who felt included now underestimated the pain they experienced when they were excluded previously.

Consequences of the empathy effect

According to Nordgren, van Harreveld, and van der Pligt (2009), as a consequence of the empathy gap effect, individuals will tend to overestimate their capacity to restrain their urges and temptations. Because of this bias, they might pursue courses of action that disregard the intensity of these impulses. They will often, therefore, be unable to restrain these urges and temptations.

This sequence of events was substantiated by Nordgren, van Harreveld, and van der Pligt (2009). In one study, fatigue was evoked in some, but not all, participants. That is, some participants completed a demanding memory task, which spanned 20 minutes and elicits fatigue. Other participants completed a memory task that is not as taxing and spans only 2 minutes.

Next, they answered questions that assess the belief they can control their fatigue. Finally, they received questions that ascertain the percentage of study time they will defer until the last week of semester.

Compared to participants who experienced fatigue, participants who did not experience fatigue were more likely to overestimate their capacity to override these feelings. They endorsed items like "I have more control over mental fatigue than the average person"--consistent with the empathy effect. Furthermore, these participants were more likely to defer their study until the last week. That is, they did consider the possibility they will be too fatigued to study most of the content in one week.

In the second study, participants were approached either before or after entering a cafe. Presumably, only participants entering the cafe were hungry. Next, participants ranked seven snacks from their least to most favorite. Finally, they were told to choose a snack. They were informed they will win money, as well as the snack they choose, if they do not consume this product over the next week.

If participants were not hungry, they were more likely to choose their favorite snack. That is, they were more likely to feel they could override the temptation to consume this product. Although not quite significant, these participants were also more likely to consume this snack and forego the reward (Nordgren, van Harreveld, & van der Pligt, 2009).

In the third study, the capacity to curb impulses was manipulated. That is, after completing a bogus implicit association test, some participants were informed they can curb impulses effectively. Other participants, after completing the same task, were informed they cannot curb impulses effectively. Next, participants, all of whom were smokers, were asked to watch a movie about cigarettes but refrain from smoking. They were informed they will be rewarded financially if they abstain from smoking a cigarette. However, the level of this reward depended on the level of temptation to which they exposed themselves. They would be paid more, for example, if they maintained an unlit cigarette in their mouth.

If participants had been informed they can readily curb their impulses, they chose a more tempting alternative. That is, they assumed they could restrain these temptations. These participants, however, were also less likely to restrain from smoking.

Limited understanding of other people

Because of the empathy gap effect, people will often not empathize with the feelings of other individuals very well. For example, in one paradigm, participants are asked to estimate the extent to which they would be willing to dance in front of a large audience, on a scale of 1 to 11. Next, they are asked to estimate the degree to which another person would be willing to dance in front of a large audience. In general, participants feel that another person would be more willing than would they to engage in this act, called the illusion of courage (Van Boven, Loewenstein, & Dunning, 2005).

Similarly, participants are sometimes asked to estimate the minimum amount of money they would need to be paid before they would dance in front of a large audience. Next, they are asked to estimate the minimum amount of money that another person would need to be paid to dance in front of a large audience. Again, participants assume that another person would be willing to dance for less money (Van Boven, Loewenstein, & Dunning, 2005).

Presumably, when individuals imagine themselves dancing, they experience the visceral reactions that correspond to embarrassment. They are, therefore, unwilling to dance in front of an audience. When they imagine someone else, they do not experience these visceral reactions: They do not identify closely enough with this person.

Consistent with this explanation, after participants read a story that includes many collective pronouns, such as we or us, this effect dissipates (Woltin, Yzerbyt, & Corneille, 2011). Presumably, these pronouns activate an interdependent self construal (see self construal theory), in which they feel connected to other people. Consequently, they identify with the other person, recognizing the visceral reactions that other people will experience. They appreciate this person may feel too embarrassed to dance in front of a large audience as well (Woltin, Yzerbyt, & Corneille, 2011).

As an aside, these collective pronouns curb the empathy gap effect only when people consider the other person and not themselves first (Woltin, Yzerbyt, & Corneille, 2011). Perhaps, if people consider themselves first, their sense of connection with the other person diminishes.

Implications of the empathy effect

The empathy effect can explain relapse. In particular, as Nordgren, van Harreveld, and van der Pligt (2009) showed, several weeks after smokers withdraw from cigarettes, their urges dissipate. As a consequence, they overestimate their capacity to control these urges. Hence, they become less inclined to avoid environments in which smoking is rife. They might even smoke one or two cigarettes, because the assume they can curb any subsequent urges.

Individuals should, therefore, formulate plans in the same state as which these intentions will be executed. They should, for example, plan their work or study timetable while fatigued& otherwise, they might overestimate their capacity to override this fatigue.

The empathy effect can also explain inadequate policies. For example, as Nordgren, Banas, and MacDonald (2010) demonstrated, when teachers do not feel excluded, they do not recommend comprehensive interventions to enhance bullying. When teachers do feel excluded, they recommend more effective interventions. They appreciate the distress that exclusion can elicit.

Inaccurate expectations about learning

Until the requisite skills are acquired, many of the products that individuals purchase cannot be utilized efficiently. For example, computers, cameras, cooking appliances, skis, sailboards, and many other goods demand considerable skill. As Billeter, Kalra, and Loewenstein (2010) showed, before individuals attempt to utilize these goods, they underestimate the duration that might be needed to develop this skill. However, after individuals first attempt to utilize these goods, they overestimate this duration.

Specifically, in general, individuals overestimate their capabilities. However, as they begin to learn a skill, a cognitive mindset is evoked (see cognitive experiential self theory). That is, individuals strive to memorize and apply a series of logical principles and rules. In this state, they assume they may still need to memorize and apply these logical principles and rules even after they have acquired the skill. They may not recognize that, once they acquire this skill, they will utilize an experiential mindset, in which performance proceeds seamlessly, effortlessly, and automatically. That is, while operating in a cognitive or deliberative mindset, the subjective experience of an experiential mindset is not fully appreciated--a manifestation of the empathy gap.

In short, when people initially grapple with a skill, they imagine they might need to devote the same level of effort and concentration in the future as well. They assume, therefore, they will not progress rapidly. They underestimate the benefits of an experiential mode and overestimate the duration that will be needed to become proficient.

Billeter, Kalra, and Loewenstein (2010) conducted a series of studies that corroborates this argument. In the first study, participants learned to trace a shape--two five point stars, one inside the other. However, they could observe only the reflection of this shape in the mirror. Furthermore, they could not touch the boundary that separated the two stars.

After they received the instructions, but before they attempted the activity, they were asked to estimate the number of errors they are likely to commit if they completed this task four times. Next, they practiced this task for two minutes before estimating the number of errors they are likely to commit again. Hence, they generated two predictions. In general, predictions before practicing underestimated the number of errors committed& predictions after practicing, however, overestimated the number of errors committed, consistent with the hypotheses. That is, their actual rate of improvement exceeded their predicted rate of improvement.

Five subsequent studies generalized these findings to other measures of performance as well as other tasks and conditions (Billeter, Kalra, & Loewenstein, 2010). For example, the same pattern of observations emerged when participants needed to predict the duration that will be needed to reach a specific level of performance. Also, even after participants were told about these biases, the same pattern was observed, but to a lesser extent. These findings were also obtained when other tasks, such as a novel keyboard, were attempted.

Because of this bias, individuals often underestimated the value or utility of products. That is, after they initially attempted one of these tasks, they were not as willing to pay money to purchase these goods. Therefore, as consumers learn a skill, companies may need to invest in greater levels of support to curb the likelihood of rejection.


Ariely, D., & Loewenstein, G. (2006). The heat of the moment: The effect of sexual arousal on sexual decision making. Journal of Behavioral Decision Making, 19, 87-98.

Badger, G. J., Bickel, W. K., Giordano, L. A., Jacobs, E. A., Loewenstein, G., & Marsch, L. (2007). Altered states: The impact of immediate craving on the valuation of current and future opioids. Journal of Health Economics, 26, 865-876.

Billeter, D., Kalra, A., & Loewenstein, G. (2010). Underpredicting learning after initial experience with a product. Journal of Consumer Research, 37, 546-555. doi: 10.1086/655862

Loewenstein, G. (1996). Out of control: Visceral influences on behavior. Organizational Behavior and Human Decision Processes, 65, 272-292.

Nordgren, L. F., Banas, K., & MacDonald, G. (2010). Empathy gaps for social pain: Why people underestimate the pain of social suffering, 100, 120-128.

Nordgren, L. F., van der Pligt, J., & van Harreveld, F. (2006). Visceral drives in retrospect: Explanations about the inaccessible past. Psychological Science, 17, 635-640.

Nordgren, L. F., van der Pligt, J., & van Harreveld, F. (2007). Evaluating Eve: Visceral states influence the evaluation of impulsive behavior. Journal of Personality and Social Psychology, 93, 75-84.

Nordgren, L. F., van Harreveld, F., & van der Pligt, J. (2009). The restraint bias: How the illusion of self-restraint promotes impulsive behavior. Psychological Science, 20, 1532-1528.

Sayette, M. A., Loewenstein, G., Griffin, K. M., & Black, J. (2008). Exploring the cold-to-hot empathy gap in smokers. Psychological Science, 19, 926-932.

Van Boven, L., Loewenstein, G., & Dunning, D. (2005). The illusion of courage in social prediction: Underestimating the impact of fear of embarrassment on other people. Organizational Behavior and Human Decision Processes, 96, 130-141.

Woltin, K., Yzerbyt, V. Y., & Corneille, O. (2011). On reducing an empathy gap: The impact of self-construal and order of judgment. British Journal of Social Psychology, 50, 553-562. doi: 10.1111/j.2044-8309.2011.02024.x

Academic Scholar?
Join our team of writers.
Write a new opinion article,
a new Psyhclopedia article review
or update a current article.
Get recognition for it.

Last Update: 7/6/2016