**4. Bottom-up and top-down routes to happiness: insights from neuroscience**

It is only recently that neurosciences and psychiatry have contributed to systematically studying happiness [56]. For a long time, research had focused on the emotional processes underlying pathologies. In the literature, terms such as "happiness," "well-being," or even "satisfaction" are often associated or confused [57]. Moreover, it is difficult to measure such variables otherwise than by using self-reported scales [54, 58]. The direct study of these dimensions is, therefore, hardly suitable for the classical methods used in neurosciences. As explained above (Section 3), the experience of positive emotions and their accumulation is a key element for happiness emergence. Seeking and experiencing positive affects corresponds to hedonia, an essential contributor to happiness and well-being [59]. Significant progress has been made in understanding the emergence of affects and positive emotions and their relationships with brain function. Studies in neuroscience focus on hedonia because the latter is correlated with and participates in, the emergence of happiness and well-being (see [60] for a proposed conceptual model). The purpose of this section is to provide a summary of current knowledge about the emergence and processing of positive emotions and affect and how they contribute to happiness since the former is often considered as a "proxy" for the latter. A classic standpoint in neuroscience consists in the attribution of the responsibility of the responses based on the perceptual

processing of emotional stimuli to subcortical regions (like amygdala, ventral pallidum, or thalamus). Therefore these "low-level" structures have been identified as supporting "bottom-up" processing. By contrast, "high-level" processes based on integrated cognitive processing of emotional stimuli are preferentially attributed to cortical structures (such as the prefrontal lobe), which are thought to underlie "top-down" processing [61–63]. This dissociation (subcortical regions = "bottom-up" process/cortical regions = "top-down" process) will be discussed in the last part of this section but has also been taken as a landmark to organize this section.

### **4.1 Specific regions associated with positive emotions**

George and collaborators (1995) were among the first to study brain activation as a function of stimulus valence in order to evaluate the specificity of brain coding for emotional valence [64]. The authors used H215O positron emission tomography (PET) to study changes in regional cerebral blood flow (rCBF) on a sample of 11 women when happy, sad, or neutral states were induced. In this study, happiness was considered as a transient emotional state induced by self-generated positive emotions via memory and by showing positive emotional faces with the instruction to feel the corresponding emotion. According to their results, transient happiness was associated with a reduction in rCBF in the right prefrontal and bilateral temporal-parietal regions. Sadness was associated with bilateral activation in the cingulate, medial prefrontal, and mesial temporal cortex but also in thalamus and putamen. These data suggest that there are different neural substrates activated depending on stimulus valence. Using the functional magnetic resonance imaging (fMRI), Pelletier and collaborators (2003) challenged this conclusion. They reported that neutral, happy, or sad states were associated with similar activation in the orbitofrontal, left medial prefrontal, and left and ventrolateral prefrontal as well as left anterior temporal pole. But, in those regions, loci of activation were situated in different sub-regions. Their conclusion is that happy or sad states activate similar brain areas but different neural circuits [65].

In a substantial review, in the framework of the Human Affectome Project, Alexander and collaborators (2021) have identified different key brain regions involved in the experience of positive emotions and affect. Cortical regions including prefrontal cortex (PFC), orbitofrontal cortex (OFC), and anterior cingulate cortex (ACC), and subcortical regions like insula or amygdala were considered. Although these regions may have different cognitive functions regarding the evaluation of emotional stimuli, the regulation of affective states, or decision making; they contribute to the emergence, processing, and maintenance of a positive emotional state and to the emergence of well-being and happiness [60]. In sum, the experience of positive affects is associated with a vast neural network distributed throughout the brain, which involves both cortical and subcortical regions. In the next part, we are discussing the specific role of subcortical regions, often associated with bottom-up processes.

### **4.2 Bottom-up influences on positive emotions**

Beyond the key brain areas identified in the processing and experience of positive emotions, hedonic hotspots have been identified in nucleus accumbens, ventral pallidum, forebrain, limbic cortical regions, or brainstem [66]. Based on Kringelbach et Berridge, these hedonic hotspots are said to cause the pleasure reaction in increasing

### *Enacting Happiness from Emotions and Moods DOI: http://dx.doi.org/10.5772/intechopen.106954*

"liking" feeling and are associated with hedonia. Experimentally studied using microinjections of opioids, endocannabinoids or other neurochemical substances, these areas measure approximately 1 cubic centimeter in humans. Hedonic hotspots do not work according to an on/off mode. For example, stimulation of hedonic hotspots by opioids in the nucleus accumbens amplifies "liking" but this same area stimulated by dopamine amplifies "wanting" reactions. However, "wanting" a salient reward does not necessarily mean that this reward is appreciated [67]. Authors suggest that "liking" is a component of pleasure in connection with a reward and can lead to a subjective feeling of happiness.

These hedonic hotspots make salient a stimulus and allow an approach motivation through the processes of "liking." Ventral pallidum is a region involved in the implementation of this motivated behavior. This subcortical structure is part of the reward system and its activation increases stimulus attractivity [68]. Ventral pallidum is essential to the production of behaviors adapted to reward but is also involved in the phenomena of drug dependence [69]. Studies in nonhuman animals also indicated that injury of ventral pallidum abolishes the pleasurable reaction to a stimulus, which corresponds to a state of anhedonia [70, 71]. Activation of hedonic hotspots and ventral pallidum participates in the emergence of pleasure through the evaluation and search for reward. Reward is simply made salient and attractive, and this evaluation is independent of "high-level" cognitive processes. These are "bottom-up" processes based on the perception and evaluation of a stimulus leading to a motivated, approach behavior. Seeking reward and pleasure is an adaptive behavior participating in the experience of positive emotions [66].

The insula is a region that makes the interface between cortical and subcortical regions, and also contributes to adaptive behavior. The links between the insula and positive emotion remain largely unknown. In a review, Uddin and collaborators (2017) reported on functioning and connectivity of this region [72]. The insula is connected to many cortical (frontal cortex, temporal and parietal lobe, or somatosensory cortices) and subcortical (hippocampus, amygdala, or thalamus) regions [72, 73]. From a functional point of view, the insula participates in the processing of viscero-somatosensory information, essential for interoception (ability to evaluate physiological activity; [74, 75]). Insula is fundamental in emotional experience and subjective feelings [75]. Emotional stimuli cause physiological changes and these bodily sensations (processed in particular by the insula) are essential to the subjective experience of emotions and decision-making [72, 76]. Through the visceral experience of emotions, the insula allows the implementation of emotional regulation strategies [77, 78]. By processing physiological information induced by the characteristics of a perceived valenced stimulus, the insula contributes to the "bottom-up" processes underlying positive emotions. On the other hand, it allows —by these connections with the frontal regions and the conscious access to visceral sensations— the cognitive processing of valenced stimuli and is involved in "top-down" processes.

### **4.3 Top-down influences on positive emotions**

From a "top-down" perspective on positive emotion processing, high-level cortical structures have received particular attention. Perhaps the most studied region is the prefrontal cortex (PFC). In this area, the experience of positive affects is lateralized in the left PFC whereas negative affect is lateralized in the right PFC [79]. Recently, Mendez and Parand (2020) documented a clinical case of a patient who suffered significant brain injury. Following a suicide attempt by gunshot, medical imaging

examinations of this 63-year-old man showed significant volume loss of the right frontal and right anterior lobes and injury in left orbitofrontal. Clinically, this brain injury led to a significant personality change. The patient was described as permanently "happy" and scoring 28/28 on the Subjective Happiness Scale (Lyubomirsky & Lepper, 1999), despite his many sequelae and the fact that he was imprisoned [80]. This clinical case suggests that the emergence of happiness could be related to the left PFC, while the right PFC would participate in regulating a positive biased default mode [81, 82]. Similarly, studies based on electroencephalography (EEG) have shown that activation of the left PFC is associated with positive affect and activation of the right PFC with negative affect [79]. Interestingly, this activation of the left PFC does not only occur when processing positive information but is also related to dispositional mood, temperament, and well-being [83]. Left PFC being associated with dispositions toward positive emotional states, it is a good candidate for a top-down moderating process of happiness.

However, this lateralization of the PFC is far from being as strict as it may seem. Indeed, bilateral activation of the orbitofrontal cortex (OFC) has been observed during hedonic processing [84]. OFC, and more specifically, the mid-anterior subregion is involved in the subjective experience of pleasure, an important component of hedonia [66, 85]. This brain region is also associated with optimism, a dispositional trait, and an optimistic explanatory style (i.e., how an individual reacts to a positive or negative event), underlying well-being [86]. OFC allows the subjective attribution of a hedonic valence to a stimulus and contributes to the emergence of judgment and hedonic-motivated decision-making [85]. Subjective access to the hedonic evaluation of a stimulus is associated with the deployment of "high-level" processing and "top-down" processes. Hornak and collaborators (2003) have shown that lesions of the OFC can cause a deficit in emotional voice or face recognition and produce changes in behavioral and subjective emotional states. Their work also emphasized the importance of the anterior cingulate cortex (ACC) and its links with the OFC in human emotions [87].

Classically, ACC is considered to be recruited in attentional control tasks (i.e., tasks largely involving "top-down" processes [88, 89]). Several studies also emphasize the importance of ACC in emotional regulation mechanisms [90–92]. Brassen and collaborators (2011) suggested that ACC activation is related to an attentional bias toward positive stimuli [93]. When an individual is in a positive emotional state, this attentional bias manifests in order to preserve this state. It is a motivated behavior involved in emotional regulation. Their work has thus indicated, in an elderly sample, that an increase in the activity of ACC (more particularly in the rostral part of the ACC) is correlated with this positivity bias but also with the feeling of well-being. On the other hand, ACC is involved in compassion, a trait associated with the production of positive emotion and happiness [86]. It is accepted that dysfunction of ACC contributes to certain cases of major depression [94]. Thus, it is assumed that through attentional control directed toward positive stimuli, the ACC contributes to the regulation and maintenance of a positive emotional state and is, therefore, a key structure in the emergence of happiness [60]. Beyond specific brain areas brain networks have been studied in relation to happiness. In particular, the default mode network (DMN) has been identified as a key component for the emergence of happiness [61]. DMN consists of low-intensity activity that can be recorded when the participant is at rest; it is inhibited when they are engaged in cognitively demanding tasks and not self-oriented. This is composed of bilateral cortical area located in the frontal, parietal, and temporal lobes [95]. The DMN includes medial prefrontal cortex, posterior cingulate

### *Enacting Happiness from Emotions and Moods DOI: http://dx.doi.org/10.5772/intechopen.106954*

cortex, and inferior parietal lobule [96–98]. This has a central role in the happiness experience [98]. Luo and collaborators (2016) demonstrated that increased DMN connectivity in PFC and cingulate cortex is associated with lower levels of happiness. Greater functional connectivity of the DMN is associated with more self-reflection based on negative content and ruminations leading to unhappiness [97]. These results are consistent with previous studies in the context of psychopathology [99]. Thus, when the networks involved in cognitive tasks are dominated by the DMN, there is a greater tendency to rumination and, therefore, to lower levels of happiness [100]. The default-mode network activation may interfere with positive task implementation by preventing the activation of neural networks involved in planning, preparing, and selecting appropriate behaviors [101].

### **4.4 Discussion**

The data gathered in this section support the idea that positive emotion emergence and processing cannot be related to one single dedicated brain area. It is rather supported by a complex network involving cortical and subcortical regions. Although much progress has been made, the processes associating the activation of this network with a subjective sense of happiness remain unclear and require further research [60]. It is also necessary to avoid any simplistic shortcut consisting in considering that the activation of a key brain area in the processing of positive emotions is enough to cause a subjective feeling of happiness. Indeed, the same brain area can process emotions of different valences through different neural networks [65].

As first mentioned in the introduction of this section, the classical approach in neuroscience is to assimilate responses based on sensory or perceptual processing of stimuli with "bottom-up" processing. While cognitive processing of stimuli is associated with "top-down" processing. Thus, "bottom-up" and "top-down" processing are often associated with subcortical and cortical regions, respectively [61–63]. This view has two main limitations. On the one hand, it has been shown that tasks that are supposed to require bottom-up or top-down processing mobilize both subcortical and cortical regions [63, 73]. This element is consistent with the state of current knowledge in the field of emotions. Relatively complex processes such as hedonic evaluation of stimuli, approach behavior *via* decision-making or emotional regulation are underpinned by many cortical and subcortical areas [60, 66, 73, 77, 102]. Consequently, viewing bottom-up and top-down processes as strictly competitive or exclusive presents limitations. An approach integrating these processes in a "circular" perspective seems more coherent. Both the perceptual characteristics of the stimuli and the elements resulting from higher-level cognitive processing should be taken into account. Moreover, a significant part of work in neuroscience is based on functional magnetic resonance imaging (fMRI). This technique allows a good spatial identification of cerebral activations but is generally less adapted in studies requiring high temporal resolution, which is offered by electrophysiology. It has been reported that the cerebral processes involved in evaluating positive stimuli take place more slowly than those that involve negative stimuli [60]. Esslen and collaborators (2004) have shown that the presentation of happiness face activates the frontal regions as well as the ACC according to different temporal segments [103].

Finally, we are aware that this literature often refers to various concepts. Wellbeing, happiness, hedonia, positive emotions, mood, or affect can, depending on the authors, be almost interchangeable notions or refer to completely different concepts. We chose to focus on studies that tried to link positive emotions to happiness despite

differences in definitions offered or positive emotion induction protocols and measurement tools used. This does not prevent us from taking a certain distance from this literature, which has not yet succeeded in providing a strict and consensual conceptual framework. It is necessary to be aware of this element when discussing the notion of happiness in neuroscience.

In neuroscience, the importance of frontal regions in the processing of positive emotions is clearly established. These are key brain areas in emerging happiness. Beyond the processing of positive affect, they are the seat of "high-level" functions such as "executive functions." It is therefore necessary to take an interest in these "highlevel" processes in order to understand their role in the emerging happiness. Finally, work in neuroscience teaches us that positive emotions, fundamental parts of subjective feeling of happiness, are supported by a large cerebral network. This network is flexible and its activation depends on both "bottom-up" and "top-down" processes.
