Behavioral Approaches and Psychophysics

FIGURE 11-1 Pain-indicative facial movements: Shown are those facial movements that are frequently displayed in the context of experimental as well as clinical pain conditions. Facial responses to pain have mostly been analyzed using the facial action coding system (FACS) which categorizes facial responses in different Action Units (AUs). Each picture illustrates a single AU. A–D: Depicts those AUs that have been shown to be the key facial activities reliably occurring during pain (“prototypical facial expressions of pain”). In addition, (E) depicts an AU that is often found to be pain indicate for patients with dementia.

It is, however, important to keep in mind that—although these key facial movements reliably occur during pain—this does not imply only one uniform facial expression of pain that can be observed at all time and in each individual [7]. Instead, the frequencies of occurrence of the single key movements during pain usually range only from 10% to 60% [23, 26]. Therefore, the likelihood that all four key facial movements occur simultaneously or in other words the likelihood that an individual experiencing pain displays the complete “prototypical expression of pain” is very low. Rather, individuals often display only parts of this subset, sometimes even blending it with a limited range of other facial movements (e.g., smiling [15, 30, 31]). We could recently show that it might be more helpful to differentiate between at least three different facial activity patterns of pain that are displayed in the context of pain and which are composed of different combinations of facial movements [26]. These were tightening of the muscles surrounding the eyes with furrowed brows and wrinkled nose (pattern I; combination of A + B + C of Fig. 11-1); furrowed brows with tightening of the muscles surrounding the eyes (pattern II; combination of A + B of Fig 11-1), and opened mouth with tightening of the muscles surrounding the eyes (pattern III; combination of B + D of Fig. 11-1). These different facial activity patterns all have one facial movement in common, namely the tightening of the muscles surrounding the eyes (AU 6_7). This facial movement is indeed the most frequent, and thus possibly the most important movement occurring during pain [7]. Moreover, the tightening of the muscles surrounding the eyes is also the one facial movement that helps to differentiate the very similar facial expressions of pain and disgust, which is crucial for clinical practice since facial expressions of pain and of disgust are very frequently mistaken for each other [29]. Last but not least, when describing how facial responses to pain look like, it is also important to mention that a considerable percentage of individuals (approximately 15–25%) do not show any visible facial responses during the experience of pain at all in the laboratory, although they do report moderate to even strong pain intensities [26]. This is important to keep in mind when judging pain based on facial expressions, because this indicates that individuals might be experiencing pain, although they do not show any pain-related facial activity and that a “stoic face” is thus not necessarily incompatible with the experience of pain [7, 26]!

In summary: How do facial expressions of pain look like? There is a subset of facial actions that are frequently displayed in the context of pain and these pain-indicative facial actions are shown in Fig. 11-1. These single facial movements are rather seldom displayed simultaneously when individuals are experiencing pain, but instead individuals most often show different combinations of these single facial movements. Most frequently, tightening of the muscles surrounding the eyes is paired with one or two of these other pain-indicative facial movements. However, it is also possible that no visible facial expression is displayed, even though an individual is experiencing pain.

Are Facial Expressions of Pain “Hardwired” or a Learned Behavior?

Given that dementia is associated with a decline in cognitive capacities, it is important to understand to which degree facial expressions of pain are “hardwired” or a learned behavior. If they are mostly a learned behavior, the possibility that facial expressions of pain are affected by the dementia-related cognitive decline seems more likely. Based on empirical findings, it is acknowledged that facial responses to pain are a mixture of biological dispositions as well as of social learning [11]. As for their biological dispositions, infants (including neonates) [7] and congenitally blind individuals [24] display similar patterns of facial movements in response to pain as sighted adults do (see also Fig. 11-1 for a list of the most frequent pain-indicative facial movements). These findings clearly suggest that the basis of the facial expressions of pain is indeed “hardwired.” This is an important conclusion especially when considering facial expressions of pain in patients with dementia. Given that the capacity to facially express pain appears mainly inborn, it seems unlikely that patients with dementia unlearn how to facially express pain across the course of cognitive decline, at least in its early phases. Therefore, facial responses to pain might remain unaltered in patients with dementia and can be used as a valid diagnostic indicator of pain. We will further discuss this topic later on (see section on dementia-related changes in facial responses).

Despite facial expressions of pain having been shown to be strongly “hardwired,” it is also acknowledged that facial responses become modifiable across early and late childhood through social learning experiences [11]. One very important modification regards the degree to which we express pain via our face. Whereas young children tend to show vigorous facial expressions of pain, older children and adults seem to have learned to effectively downregulate their facial expressions of pain [35]. In line with this finding, a recent functional imaging study could show that a low degree of facial expressiveness to pain was associated with higher activation in frontostriatal structures [23]. Given that these frontostriatal structures are known to be involved in motor inhibition, this finding suggests that low-expressive individuals actively suppress their facial expression of pain [23]. When trying to interpret these findings, it has been argued that individuals learn to suppress the facial display of negative affect (including pain) following culturally/socially learned “display rules.” These display rules represent social norms about when, where, and how one should express affective states [9] and are learned already at a young age. Based on this theory, facially responding to pain would be the “default” that individuals learn to suppress due to social/cultural demands (e.g., “big boys don’t cry,” “one mustn’t be oversensitive to pain”). In accordance with this theory, it has been demonstrated in previous studies that social learning and the social context do indeed influence the degree of facial expressiveness to pain. The presence of others can reduce (e.g., when being together with a stranger) as well as increase (when being together with a loved one) the amount of pain-indicative facial responses depending on the nature of the relationship between interactants [21, 50]. Furthermore, it has also been shown that the degree of facial expressions due to pain can be effectively modulated by operant learning, with facial responses increasing or decreasing dependent on the contingency of reinforcement [32].

In summary: Are facial expressions of pain “hardwired” or a learned behavior? The basic algorithms of facial expressions of pain (determining which facial movements are displayed in the context of pain) seem to be “hardwired”. This is a crucial prerequisite in order to reliably communicate one’s affective state across developmental stages, ranging from neonatal stages up to older age, as well as largely independent of one’s cognitive status. The degree to which we communicate pain via our face, however, seems to be also affected by social learning, with individuals mainly learning to downregulate their facial expressions of pain when inappropriate.

Are Facial Expressions of Pain Changed in Patients with Dementia?

So far, only a few studies have investigated facial responses to pain in patients with dementia [12–14, 27, 33, 37, 42]. Facial responses during experimental pain induction (pressure and electrical stimulation [27, 33, 37], flu injections [12], venipuncture [13, 42], and exacerbation of chronic musculoskeletal pain during physical exercise [14]) were videotaped and later analyzed using the FACS [7]. Although these studies differed immensely with regard to the way of pain induction, the findings are surprisingly homogeneous. First of all, it was consistently found that facial responses to increasing pain intensified in patients with dementia [27, 33, 37]. This means that facial expressions still encode the intensity of pain in patients with dementia. Moreover, even patients who were verbally compromised still facially displayed pain properly [27, 33, 42]. Furthermore, these facial expressions were composed of the same types of facial movements as can be found in elderly individuals without dementia in response to pain [33]. There is only one facial action that was displayed much more frequently in patients with dementia, namely the opening of the mouth (AU25_26_27) [33]. Given its high frequency of occurrence in patients with dementia we also included this facial movement in Fig. 11-1 (last row). Although opening of the mouth is not listed among the prototypical facial movements occurring during pain, this facial movement has nevertheless been frequently reported in the context of pain even in cognitively unimpaired individuals [7]. Thus, opening of the mouth is not an atypical facial response to pain, but this facial response seems to be especially pain-indicative in patients with dementia.

The similarity in types of facial movements being displayed by patients with dementia and healthy controls is illustrated in Fig. 11-2. Here, examples of facial expressions are shown that were displayed in response to painful pressure stimulation. When comparing these expressions to the pain-indicative facial responses shown in Fig. 11-1, it should become evident that the “pain typicality” of facial expressions of pain is not reduced in patients with dementia.

These findings are very promising, given that they clearly suggest that the face seems to specifically encode the experience of pain and that this specific encoding does not change in the course of dementia. Considering the degree of specific facial expressiveness during pain, facial responses in patients with dementia seem to be augmented compared to responses in cognitively unimpaired controls [14, 27, 33, 42

Only gold members can continue reading. Log In or Register to continue