Behavioral and neural correlates of pain-related language

The aim of this doctoral research project is to investigate the psycholinguistic, behavioural and neural correlates of pain-related language in healthy subjects. Specifically, we employed psycholinguistic, behavioural and brain imaging approaches in order to study (1) how pain-related language is semantically structured and organized, (2) how reading pain-related words affects the aversive motor response and (3) how reading pain-related words modulates brain activity compared to noxious stimuli. In the present dissertation we will discuss data from the psycholinguistic study and the behavioural experiments. Data from the neuroimaging experiment are still under analysis and will not be reported. The International Association for the Study of Pain (IASP) defines pain as “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage” (Bonica, 1979; Merskey, 1964). The IASP definition reveals that pain is not a mere physical sensation, but it is the subjective result of sensory-discriminative, motivational-affective, and cognitive-evaluative dimensions. These dimensions are represented in separate nodes of a complex neural network referred to as the “pain matrix” (Avenanti & Aglioti, 2006; Price, 2000). As a subjective experience, pain perception may be modulated by psychological variables such as cognitive and emotional states (Friederich et al., 2001; Johnson et al., 1991; Keltner et al., 2006; Miltner et al., 1999; Miltner et al., 1989; Price, 2000; Rainville et al., 2005) as well as more complex experiences like pain anticipation and empathy, which incorporate both emotional and cognitive factors (Anelli et al., 2012; Apkarian et al., 2005; Melzack, 1999; Morrison et al., 2007). These psychological variables and complex experiences may modulate pain perception for better or for worse, i.e., may decrease pain perception or increase it. Moreover, a number of different stimuli, when associated with pain, e.g., faces, pictures and videos, can activate the pain matrix, or part of it, even in absence of a noxious peripheral stimulation. Only a few studies have employed words related to pain to investigate their effect on pain perception. Pain-related words were found to alter pain itself and to activate brain structures engaged in the processing of noxious stimuli (Ritter et al., 2016). Pain is often one of the main complaints of patients. As a subjective experience, its evaluation must be a relevant element in the outcome assessment, which is a critical step to provide good pain management. Indeed, language is a means to create expectation of pain or pain relief. The pain evaluation in clinical setting has been dominated by the paradigm introduced by the McGill Pain Questionnaire (MPQ: Melzack, 1975), a word-based instrument for the multidimensional assessment of pain. The MPQ has been widely used in the research on pain as well and has significantly stimulated a large body of human research and methods of pain evaluation. Nevertheless, its predictive value on pain states remains to be demonstrated (Main, 2016). Moreover, its 78 pain descriptors do not report values for any of the psycholinguistic and emotional variables that are now known to affect comprehension processes. Pain is a negative affective state. Its aversiveness, i.e., its negative affective-related nature, is important for self-preservation by promoting withdrawal from harmful or potentially harmful stimuli (Navratilova & Porreca, 2014). A fundamental assumption in emotion research is that emotions predispose the organism to act adaptively in a frequently changing environment (Kozlik et al., 2015). Many emotion theories postulate the existence of a motivational orientation, i.e., a link between emotion and action tendencies that prepare the organism to act adaptively. Specifically, positive stimuli were assumed to activate approach motivational circuits, which in turn trigger approach-related behavioral tendencies, whereas negative stimuli, such as pain-related stimuli, should activate aversive motivational circuits, which trigger avoidance-related behavioral tendencies (Kozlik et al., 2015). It is possible that the evolutionary relevance and the salience of pain-related stimuli could differently affect the avoidance-related behavior compared to negative, pain-unrelated stimuli. Interestingly, words used to describe physical pain-related experiences are also used to describe social pain, i.e., the unavoidable feeling of pain associated with being socially rejected, excluded or losing those closest to us (Baumeister & Leary, 1995; MacDonald & Leary, 2005; Eisenberger, 2015). From an evolutionary perspective, social bonds are a fundamental need and, like other basic needs, their lack represent a threat to survival and physical safety. Neurochemical, neurophysiological and brain imaging studies support the hypothesis that social pain processing might relay on the same mechanisms that process physical pain and that losing social connections may lead to a painful perception, allowing people to describe this experience with words typically used to report physical pain (e.g., a broken heart, a soul scars). Although such studies have involved the use of a number of different kinds of stimuli to induce social pain, e.g., pictures, memories, painting, no studies about the effects of social pain-related words have been carried out so far. In the first part of this dissertation, theoretical and empirical background of the experimental questions investigated is described. This introductory part includes three chapters. In the first chapter we provide a description of the pain system, from nociceptors decoding noxious insults to cortical and subcortical processing of pain perception, and of the modulatory effect exerted by nocebo. We also outline the issue of the pain measurement in clinical setting. In the second chapter we review studies about the effect that pain-related words have on pain processing. In the third chapter, we describe what social pain is and why it is thought to relay on pain matrix. In the second part of the dissertation we describe the studies we carried out. Study 1. The first study was designed to investigate how pain-related language is semantically structured and organized. We employed an online, large-scale survey to collect data about several psycholinguistic and emotional variables that affect the recall of their semantic meaning Our hypothesis was that the semantic content of pain-related language differentiates its psycholinguistic and emotional structure from general emotional language. When it comes to study pain, psycholinguistic and emotional pain-related word databases are available neither in Italian nor in any other languages. This study also aims to fill this gap by creating a dataset of pain-related words with ratings for psycholinguistic and emotional variables available for research purposes. We also provided scores of words’ pain-relatedness, i.e., the subjective estimate of the degree to which the word is generally linked to pain semantic domain, as well as the intensity and the unpleasantness they convey. Results showed that the semantic structure of pain-related language is partly different from the semantic structure of generic language, mainly because the modulatory effect of pain-related variables. The second study includes three behavioural experiments designed to assess approach/avoidance tendencies toward pain-related linguistic stimuli in a healthy population. Experiment 1. In this experiment we assessed if pain-related words, given their individual saliency and relevance, produced a different patter of results, with larger avoidance responses, than negative pain-unrelated words. We set up the experiment using an object-related reference that allows considering the withdrawal of the hand as avoidance and its movement on as approach. This frame would reproduce the adaptive automatic behavior that individuals experience with noxious stimuli. We asked participants to perform an explicit task to assess if the semantic content of pain words and its evolutionary relevance might affect the response when an explicit evaluation of the valence is required. We recorded the RTs of the movement initiation (release RTs) as well as the time requested to perform the complete arm movement (pression RTs). Results showed that pain-relatedness affects movements independently by the motivational orientation and does not trigger automatically approach/avoidance behaviors. Interestingly, social pain words exerted the strongest effect. A negative linear relationship between pain-related variables and RTs performing the movement emerged, regardless the direction of the movement itself. Experiment 2. This experiment compared the implicit task to the explicit task of the Experiment 1. Participants performed a lexical decision on the same stimuli and with the same paradigm used in Experiment 1. Interestingly, the different, implicit task triggered a congruent approach/avoidance behavior. Again, social pain words exerted the strongest effect. However, pain affects pression RTs in an unspecific manner. Experiment 3. This experiment involved the use of the Manikin Task in order to disambiguate distance-change (Krieglmeyer et al., 2010) in order to test effects on approach-avoidance responses independently of labeling the motor output in terms of approach-avoidance. Subjects had to move a manikin toward or far away a valenced word and to do it, the manikin must be moved either up or down, dependent by its position related to the word. Data confirm that the Manikin Task allows the stimulus valence to facilitate compatible approach-avoidance responses even though participants had no intention to approach or to avoid and the valence of the response labels was dissociated from the approach-avoidance movement. However, differences between negative, pain-unrelated and pain words do not emerged