3.1. Evidence from neuroanatomy
Neuroanatomic evidence emphasizes the relevance of an “interoceptive neural network” in the brain comprising the somatosensory and somatomotor cortices, the insular cortex, cingulate cortex (ACC), and prefrontal cortices (ventromedial prefrontal cortex, dorsolateral prefrontal cortex). These structures are relevant for monitoring the internal emotional and viscerosensory state (Critchley, Corfield, Chandler, Mathias, & Dolan, 2000; Critchley et al., 2003), for emotion processing and reactivity (Phan, Wager, Taylor, & Liberzon, 2002), for the feeling of self-generated and externally induced emotions (Anders et al., 2004), and the self-regulation of feelings and behavior (Beauregard, Levesque, & Bourgouin, 2001; Bechara, 2004).
Within this interoceptive network the insula represents a relevant projection site of viscerosensory input from different modalities from the body. Interoceptive stimuli that activate the anterior insular cortex (AIC) include thirst, dyspnea, the Valsalva manoeuver, “air hunger,” sensual touch, itch, heartbeat, and distension of the bladder, stomach, or esophagus (see Craig, 2009b). Evidence from work on recognition of body movement, music and rhythm, emotional awareness, self-recognition, and time perception underscores the relevance of the insular cortex and/or the ACC (Craig, 2009b) for all subjective human feelings.
Findings on the neural basis of interoception by using heartbeat perception confirm the relevance of the interoceptive neural network and show that good compared to poor heartbeat perceivers demonstrate greater activation, especially in the right AIC (Critchley et al., 2004; Pollatos et al., 2007c). The weight of evidence suggests greater central representation and integration of cardiovascular signals in persons who are more aware of their cardiac signals. A recent model of interoception (Craig, 2009b) underscores the role of the insula for the translation of visceral and further bodily states into subjective feelings and self-awareness. Accumulating evidence (Craig, 2008, 2009a,b) highlights the relevance of a phylogenetically new homeostatic afferent lamina-1 spinothalamocortical pathway that converges to “interoceptive centers” in the insular and orbitofrontal cortices (Craig, 2009a,b) and gives rise to conscious visceral perception (see Fig. 1).
Figure 1. Suggested posterior-to-anterior progression in the insula (VMPFC, ventromedial prefrontal cortex; DLPFC, dorsolateral prefrontal cortex; figure adapted from Craig, 2009a).
Download figure to PowerPoint
It is suggested that different portions of the insula are involved in different and successive steps of neural processing building the basis of the sequential integration of the primary homeostatic condition of the body with salient features of the sensory environment and with motivational, hedonic, and social conditions. Raw interoceptive signals such as those coming from visceral changes and pain, first project to the posterior insula and become progressively integrated with contextual motivational and hedonic information as they progress toward the anterior insula. The neural constructs of the distinct, individually mapped feelings in the posterior insular cortex are then re-represented in the mid-insula that integrates the homeostatic re-representations with activity associated with emotionally salient environmental stimuli of many sensory modalities from different parts of the brain.
The culmination point of this progression is a complex and rich representation of the “global emotional moment” in the anterior insula that represents the ultimate representation of all one’s feelings, thereby constituting the “sentient self,” in the immediate present (now). In this model, the AIC provides a unique neural substrate that instantiates all subjective feelings from the body and feelings of emotion at a certain moment of time supporting the proposition that subjective awareness is built on homeostasis. In this view, the neural basis for “awareness” is the neural representation of the physiological condition of the body, and the homeostatic neural construct for a feeling from the body is the foundation for the encoding of all feelings (Craig, 2009b).
This is in accordance with the “somatic marker” hypothesis of Damasio (1994, 1999) stating that this meta-representation of bodily states constitutes an emotional feeling, accessible to consciousness and providing the “gut-feeling” that guides our decision processes and forms the basis for our “self” and consciousness. The neuroanatomic basis of interoception represents the link for the “body in the mind” and the mechanisms of the embodiment of affective and cognitive functions.
3.2. The role of interoception for decision making, emotions, and behavior: Signs of embodiment
William James (1884) stated that the experience of emotion could be defined as the perception of bodily responses and following models (Craig, 2009b; Damasio, 1994, 1999) suggest that the foundation for our emotional feelings lies in the neural representation of the physiological condition of the body, with “somatic markers” evoking feeling states that influence cognition and behavior. Theories of embodied cognition hold that higher cognitive processes operate on perceptual symbols and that concept use involves reactivations of the sensory-motor states that occur during experience with the world (e.g., Niedenthal, 2007). Similarly, activation of interoceptive representations and meta-representations of bodily signals supporting IA are profoundly associated with emotional experience and cognitive functions. A recent study performed by Tsakiris and colleagues  revealed that IA modulated the integration of multi-sensory body-percepts, thus highlighting the important role of interoception and IA for multi-sensory integration.
There is ample evidence suggesting that IA is crucial for the intensity of emotional experience (e.g., Barrett, Quigley, Bliss-Moreau, & Aronson, 2004; Herbert, Herbert, and Pollatos, 2007a, 2010a,b; Pollatos, Gramann, & Schandry, 2007a; Pollatos, Kirsch, & Schandry, 2005; Wiens, 2005) and the higher order processing of emotional stimuli (Herbert et al., 2007a; Pollatos et al., 2005). Studies employing decision-making tasks and risk manipulation provide evidence that deficits in the generation and/or representation and processing of physiological arousal are profoundly associated with disadvantageous and more risky decision behavior (Bechara, 2004; Bechara, Damasio, Damasio, & Anderson, 1994; North & O’Carroll, 2001). Recent data confirm that the accuracy with which bodily, cardiac signals are perceived is associated with benefits in decision making (Werner, Jung, Duschek, & Schandry, 2009). Moreover, IA has been shown to constitute a decisive factor for the behavioral self-regulation in situations that allow for the self-control of behavior such as physical workload (Herbert, Ulbrich, & Schandry, 2007b). These findings indicate that IA is crucially associated with the self-regulation of behavior in different situations of daily living that are accompanied by somatic and/or physiological changes giving rise to “somatic markers.” The relevance of our “gut-feelings” for decision making and behavior especially shows up in situations of uncertainty and complexity where we are free to decide upon our own actions (Damasio, 1994). Furthermore, the importance of interoceptive brain structures, especially the AIC, has been shown for risk prediction (Preuschoff, Quartz, & Bossaerts, 2008) and feelings of anticipated value during purchase and sales decisions (Knutson, Rick, Wimmer, Prelec, & Loewenstein, 2007). These insights have become a relevant part of research in neuroeconomics (Loewenstein, Rick, & Cohen, 2008).
Further results report that cardiac awareness is positively associated with benefits in selective and divided attention (Matthias et al., 2009), suggesting greater IA to represent an indicator of greater attention allocated toward both internal and external relevant events as well as self-focused attention. These results highlight the “visceral” embodiment of emotional and cognitive processes. Taken together, convincing empirical evidence and theoretical foundations have been provided for the relevance of interoception for feelings and cognitive functions. However, it should be borne in mind that most of the findings up to now are based on correlational data and imply the objection to ask in how far interoception is indeed causally involved in our emotional and cognitive processes.
In order to answer this question, future studies are necessary to manipulate interoceptive signal processing under experimentally controlled conditions and to investigate its effects on emotional and cognitive functions. There is initial evidence showing that the training of heartbeat perception or experimentally evoked changes in autonomic nervous system activity and associated cardiodynamic functions induce an improvement of the perception of internal cardiac signals. This improvement is in turn related to an enhancement of brain functions reflecting cardiac signal processing as well as to an increase in emotional experience (Schandry & Weitkunat, 1990; Schandry et al., 1993).
3.3. Embodiment of time perception
The perception of time is part of human experience; it is essential for everyday behavior and for understanding any kind of complex behavior. Recent debate connects time perception with bodily responses and embodiment by assuming that physiological states and emotions associated with changes in physiological states underlie our perception of time (Craig, 2009b; Wittmann, 2009). Such a direct link between the perception of time and physiological processes has been proposed by Craig (2009a) who claims that our experience of time relates to emotional and visceral processes because they share a common underlying neural system, the insular cortex and the interoceptive system. Wittmann (2009) follows that since emotions and physiological states seem so fundamental to the experience of time, it is tempting to assign a pivotal role to these processes related to a core timekeeping system. In line with these hypotheses, it is conceivable that the number and rate of body signals accumulated in the insula over a given timespan create our perception of duration.
In a recent study Wittman and co-authors (Wittmann, Simmons, Aron, & Paulus, 2010) found empirical evidence for this assumption: They demonstrated that during the encoding of time intervals (9 and 18 s tone intervals) activation curves over time show an accumulating pattern of neural activity, which peaks at the end of the encoding interval within bilateral posterior insula and superior temporal cortex. They argue that the accumulation function in the posterior insula might be correlated with the encoding of time intervals as suggested by Craig (2009a,b). Craig’s model proposes a close interaction between interoceptive processes and time perception, suggesting that our experience of time emerges from emotional and visceral states processed in the insular cortex. It can be interpreted as a series of global emotional moments that are indexed across a finite period of present time, from the past into the anticipated future. These series produce a cinemascopic “image” of the sentient self that is continuous across a moving window of present time. Across any point of such consecutive processing hierarchies problems can occur. In the following we will try to highlight some examples of disturbances in embodiment that are very clearly associated with abnormalities in interoceptive functioning.