NutritionNerd
Confident
This article covers scientific research done on many "alternative" remedies which are of interest to PTSD sufferers. I had to split the article to get it to fit.
Review
Sensorimotor modulation of mood and depression: An integrative review
R. Canbeyli Corresponding Author Contact Information, E-mail The Corresponding Author
Psychobiology Laboratory, Department of Psychology, Bogazici University, 34342 Bebek, Istanbul, Turkey
Received 18 August 2009; revised 29 October 2009; Accepted 2 November 2009. Available online 11 November 2009.
Abstract
Several lines of research on mood disorders reveal that depression involves a dysfunction in an affective fronto-limbic circuitry that involves the prefrontal cortices, the cingulate cortex, several limbic structures including the amygdala and the hippocampus, lower brainstem structures and the basal ganglia. In dealing with both depressive symptoms as well as their manifestation in the brain, clinical as well as basic research has emphasized mainly a top-down or central approach in elucidating the etiology of depression and its therapy. The present integrative review emphasizes the bottom-up or peripheral view of evaluating the impact of stimulation via sensory modalities or the motor system on the same circuitry in effecting mood regulation and possibly causing mood disorders, specifically depression. The paper shows that there is now a considerable accumulation of data from clinical observations as well as research with animal models to suggest that hypo- or hyper-activation of the sensory or the motor systems by manipulating visual, auditory, olfactory or gustatory inputs as well as physical exercise can have modulatory effects on mood and depressive symptoms. Moreover, depression in turn affects sensorimotor processing, resulting in an interaction that may further contribute to the aggravation of depressive symptoms. The paper also cites evidence that activation of the affective circuitry by central manipulations such as by means of deep brain stimulation has similar modulatory effects as peripheral stimulation on mood and depression. Finally, it is proposed that systematic investigations using animal models of depression on the impact of uni- or multisensory manipulation as well as of physical exercise may provide new insights into the etiology and treatment of depression in humans.
Keywords: Depression; Mood; Mood disorders; Peripheral sensory activation; Exercise
Article Outline
1. Introduction
2. Vision
2.1. Light alleviates depressive symptoms
2.2. Alleviation of depressive symptoms in animals
2.3. Inadequate light photoreception aggravates depressive symptoms
2.4. Animal studies
2.5. Emotion and mood can affect visual processing
2.6. Neuroanatomy
2.7. Regulation of mood by visual activity in the fronto-limbic circuitry
2.8. Mood regulatory action of light via the suprachiasmatic nucleus
3. Audition
3.1. Auditory stimulation modulates mood
3.2. Studies with animals
3.3. Hearing impairment affects mood
3.4. Depression affects auditory processing
3.5. Neuroanatomy
4. Olfaction <Aromatherapy>
4.1. Modulation of mood by means of odorants
4.2. Studies with animals
4.3. Olfaction in depression
4.4. Neuroanatomy
4.5. Olfactory bulbectomy
5. Taste
5.1. Mood induction by tastants
5.2. Effects of depression on taste and food intake
5.3. Altered appetite in depression
5.4. Increased appetite and taste
5.5. Decreased appetite and anhedonia
5.6. Neuroanatomy
6. Multisensory stimulation
6.1. Peripheral stimulation
6.2. Stress and depression
6.3. Enriched environment
6.4. Central stimulation
6.5. Transcranial magnetic stimulation (TMS)
6.6. Vagal nerve stimulation (VNS)
6.7. Deep brain stimulation (DBS)
7. Exercise
7.1. Exercise improves mood and alleviates depression
7.2. Lack of exercise or exercise withdrawal induces depressive symptoms
7.3. Animal research
7.4. Neuroanatomy
8. Conclusion
8.1. Sensorimotor regulation of mood acts via its over- or under-activation of a central circuitry
8.2. The bidirectional nature of sensorimotor manipulation of mood may provide new insight into depression
8.3. Crossmodal interaction of sensory inputs on mood regulation may be relevant to mood regulation
8.4. Neuroanatomical considerations
8.5. Cellular considerations
Acknowledgments
References
1. Introduction
Major depression, second only to hypertension as the most frequent illness not only has severe consequences for the affected individuals but also to their families and social circles. Despite a long history of clinical investigations and basic research into the etiology of depression, the neurobiological basis of mood disorders is still unknown and successful outcome of treatment either in the form of psychotherapy and/or pharmacothreapeutics is far from certain [136] and [263]. Research on the neurobiology of mood disorders has uncovered no biomarker pathognomonic to depression; in fact evidence strongly suggests that depression is not due to a focal dysfunction in the brain. Neuroimaging studies of depressed patients as well as research with animal models indicate that depression is likely to arise out of a dysfunction in a circuitry comprising several cortical and subcortical structures. Observations based on neuroimaging studies with depressed patients and lesion-induced or secondary depression due to neuropathologies such as Parkinson's disease as well as studies with animal models have begun to converge on a fronto-limbic circuitry as the potential basis for mood regulation and mood disorders [78], [113], [114], [143], [214], [218], [236], [237], [245] and [307]. The main cortical components of this circuitry include the ventromedial and dorsal prefrontal cortices (VMPFC and DPFC) and the anterior cingulate cortex. Together with these cortical structures, and forming an extensive fronto-limbic circuit are subcortical structures that include the amygdala, the hippocampus, the hypothalamus as well as the raphe nuclei, nucleus accumbens and the locus coeruleus. The basal ganglia form an integral part of the circuitry not only because of their involvement in psychomotor impairments in depression, but also because of their direct contribution to an affective-motor system [5], [6], [292] and [293].
In light of the circuitry depicted above, and in accordance with the view that depression is a dysfunction of a system rather than of a focal central mechanism, clinical applications as well as basic research on the etiology and treatment of depression have emphasized the complex affective and cognitive situations that figure in mood disorders. With this ‘top-down’ approach, precipitation of depression in humans and induction of depressive symptoms in animals have been considered in light of and attributed to configurations of stressful situations [193], [194], [283] and [305]. Thus, depression is considered an outcome of central dysfunction induced by such stressful situations, undoubtedly affected by genetic factors [84] and [213], reflected as depressive symptoms in behavior and cognitive functions [186] and [187]. While this approach has produced many new insights into and possible therapeutic tools for mood disorders, the present integrative review aims to show that a diametrically opposite approach is not only possible, but may also be productive in understanding and possibly treating mood disorders. Specifically, the present paper takes as its starting point the extensive accumulation of data on the impact of manipulating sensory modalities and the motor system on both the genesis and therapy of mood disorders. As will be shown in subsequent sections, a wealth of findings culled from clinical data as well as studies with animals suggest that, depending on the intensity and duration, modulation of sensory input or motor activity can have ameliorative or debilitating effects on mood. The present paper, therefore, starts from a bottom-up approach by considering the contribution of unisensory as well as multisensory inputs to mood disorders. Additionally, the model considers the motor/behavioral side of the underlying mechanism not as a mere output conduit for psychomotor symptoms of depression but as possible cause of and potential basis for treatment for such symptoms.
The proposed bottom-up approach to mood is predicated on the hypothesis that hypo- or hyper-activation of a sensory modality or of the motor system can have a modulatory effect on mood in general and depression in particular. The proposed approach also emphasizes the bidirectional nature of sensorimotor activation and depression, considering not only sensorimotor manipulation of mood, but also the effect of depression on sensorimotor processing. It will be argued that the interactive nature of sensorimotor activation and depression is not only a major complication in the clinical setting, but the realization of its implications may actually open new avenues of animal research on depression and eventual insights into both the etiology and treatment of mood disorders. While there is as yet no clear assessment of the relationship between mood regulation and the level of sensorimotor activation, the present proposal will suggest that there is adequate empirical evidence to warrant a systematic approach to the psychophysical evaluation of the impact of the level of sensorimotor activation on mood regulation and depression.
The bottom-up approach to mood regulation proposed in the present paper proposes that the activation of an ‘affective-motor’ circuitry by sensory stimulation or physical activity is capable of modulating mood. In support of this view, it will be shown in one of the following sections that central as well as peripheral activation of the proposed circuitry does modulate mood and affect depressive symptoms. The present review does not aim to provide a detailed account of the involvement of the extended neural circuitry in the proposed model. There are several reviews that address this issue directly and show, among many techniques, that neuroimaging studies indicate changed neuroanatomical characteristics of the structures involved in the affective circuitry (for a recent review see for example Ref. [176]). However, each section below will point to studies that demonstrate a relationship between the specific sensorimotor system discussed and the neuroanatomical structures in the proposed circuitry that figure in mood modulation in that modality.
The present paper is not meant to be an exhaustive assessment of sensorimotor manipulations relevant to mood and its disorders, but as an integrative evaluation aimed to emphasize the relevance of such manipulations to mood when systematic/parametric evaluation of such manipulations have been made (for this reason, for instance, the sense of touch is not explicitly discussed because of the rather informal/unsystematic manner in which tactile stimulation has been applied as an adjunct treatment). Likewise, discussion is limited to unipolar depression.
2.0 Vision
In addition to its all important role in providing sight through the visual system, light can affect mood in both humans and animals. The modulatory nature of the relationship between photoreception and mood is shown by the fact that photic stimulation can alleviate, whereas impaired vision or inadequate light reception can aggravate depressive symptoms.
2.1. Light alleviates depressive symptoms
Regulation of mood by light is a universally known phenomenon experienced by many around the globe as a result of short or long days in the winter and summer, respectively. In contrast to the other sensory modalities to be discussed in the present paper, affective impact of photic stimulation has been investigated systematically so that light therapy is now a widely used therapeutic tool for the treatment of depression. Starting in 1980s light therapy in the form of exposure to a bright source of light for approximately an hour either in the morning or early evening has been used successfully to alleviate symptoms of both seasonal [10], [81], [105], [167], [168], [221], [257], [258], [326] and [327] and non-seasonal depression [88], [151] and [181]. There is also evidence that exposure to artificial bright light can have ameliorative effect on mood in nonclinical populations. A study conducted in Finland in the winter indicated that repeated exposure to at least 1 h of light in office workers reduced depressive symptoms in healthy subjects with or without winter-related mood symptoms [224]. Interestingly, another Finnish study with nonclinical subjects indicated that bright light exposure (2500–4000 lx at eye level) paired with exercise over an 8-week period significantly reduced depressive symptoms compared to exercise in normal ambient illumination (400–600 lx) [166].
2.2. Alleviation of depressive symptoms in animals
There are a few studies that indicate that exposure to light, particularly in the form of long photoperiods has ameliorative effect on depressive symptoms in animals. Prolonged exposure of male rats to a long photoperiod (15 h:9 h light/dark cycle) for 30 days has an antidepressant effect as measured by reduced immobility in a forced swim test [203]. The ameliorative effect of light treatment was comparable to 30 days of treatment with the antidepressants clomipramine or desipramine. There are also indications that a much shorter exposure to light exposure can have protective effect on depression in the rat. Yilmaz et al. [338] showed that a single 12-h light exposure in female rats maintained normally on a 12 h:12 h L/D cycle has antidepressant effect as measured by behavioral despair. Subsequent studies have shown that even shorter light pulses of 30 min [278] or 10 min duration [124] delivered late in the dark phase of an L/D cycle have ameliorative effect in behavioral despair, an animal model of depression [241].
2.3. Inadequate light photoreception aggravates depressive symptoms
Just as enhanced photic stimulation is conducive to good mood and can alleviate depressive symptoms, low levels of illumination have the opposite effect of leading to negative mood and aggravating depressive symptoms [87], [125] and [152] and depression is more prevalent in areas receiving little sunlight throughout the year [29]. Given these findings, it can be expected that people with visual impairment that reduces the level of experienced illumination should show more depressive symptoms. This is in fact confirmed by clinical studies that report significant rates of depressive symptoms in patients with ophthalmic dysfunctions that compromise photoreception and/or reduce visual acuity [260], [261] and [285]. Within the framework of the sensorimotor model presented here, it should be emphasized that another factor contributing to the increased depressive symptoms in people with compromised photoreception, particularly in older subjects, is reduced motility due to impaired vision. In fact, a study that explicitly investigated activity levels along with photoreception in age-related macular degeneration (AMD) indicated that activity loss due to vision problems was a mediating factor in the relationship between visual acuity and depressive symptoms [259].
2.4. Animal studies
A 3-week exposure to a short photoperiod (5 h light/9 h dark cycle) induces depression-like behavior in sand rats compared to controls maintained on a 12 h light/12 h dark cycle [85]. Compared to animals maintained on a long-day regimen (16 h light/8 h dark days), similar depression-like results have been reported in male but not female Siberian hamsters that were exposed for 3 weeks to short days (8 h light/16 h dark cycle) starting at the time of weaning [244]. Another study with Siberian hamsters indicated that perinatal exposure to short days (8 h light/day) for 20–23 days after birth leads in adulthood to more depressive-like behavior in forced swimming compared to exposure to long days (16 h light/8 h dark cycle) [247].
2.5. Emotion and mood can affect visual processing
Just as visual input affects mood, emotions and depression can in turn modulate visual processing. Recent neuroimaging studies indicate that emotions can affect central visual processing even at the earliest stages of processing [3], [14] and [195]. A few studies have also begun to show that ‘emotional vision’ is particularly affected by depression. An fMRI study with depressed (MDD) patients, for example, showed reduced activity in the right anterior cingulate cortex and the left insula to positive pictures and in the right insula and hippocampus to negative pictures [162]. There was also a significant correlation between severity of depression and activity induced by negative pictures in the left amygdala, the left insula and bilateral inferior orbitofrontal cortex. Another study comparing event-related potentials (ERPs) to colors and emotional slides in MDD patients and healthy subjects found reduced responses in late aspects of visual processing in MDD patients (decreased amplitudes of P3 and pSW) which recovered after pharmacotherapy [227].
2.6. Neuroanatomy
Light can affect mood via two major pathways. In its non-visual function, light cannot only affect the proposed limbic circuitry via the amygdala (by thalamic relay) and frontal cortical pathways, it can also modulate mood by its action on the suprachiasmatic nucleus (SCN) that regulates many biological rhythms on a daily and yearly basis.
2.7. Regulation of mood by visual activity in the fronto-limbic circuitry
Visual input can modulate mood by a cascade of events that starts with the visual cortices, continuing in the dorsal and the ventral visual projections involving the parietal and inferotemporal cortices, respectively. The ventral stream in turn projects to the fronto-limbic structures involved in affective processing, namely the orbitofrontal cortex, the ventromedial prefrontal cortex, the cingulate gyrus, the hippocampus and the amygdala. These structures, particularly the amygdala, can induce depressive symptoms via their to connections with the hypothalamus, the basal ganglia and brainstem structures including the locus coeruleus and the raphe nuclei [48], [49], [150], [161], [172], [250] and [314]. The amygdala figures prominently in affective processing of visual input because of its connections with many cortical structures involved in visual processing including the striate cortex, the temporal cortex, and the frontal lobes as well as subcortical structures implicated in depression [160] and [235]. The fact that fear in a face can be detected by people with damage to the striate cortex (those with ‘blind sight’) [67] but not by those with damage to the amygdala [319] attests not only to the critical role of the amygdala in emotional processing, but also indicates that emotional visual processing can occur without conscious awareness and without the detailed processing that is achieved by the visual cortical areas.
2.8. Mood regulatory action of light via the suprachiasmatic nucleus
In addition to the circuitry depicted above, light input can modulate mood via its effect on the SCN, the master oscillator in the hypothalamus that regulates many circadian as well as circannual biorhythms. Briefly, light acts a zeitgeber to entrain SCN so that the large number of biological rhythms are eurythmic throughout the day. Moreover, the relative lengths of the day (light) and night (darkness) portions of the day show regular changes throughout the year, thereby providing the basis for the biological clock to regulate rhythms on a yearly basis [206], [207] and [315]. While social cues and other factors can also entrain the SCN, light is the primary input to the biological clock. Information about light is sent to the SCN via a direct retinohypothalamic tract, or indirectly via projections from the intergeniculate leaflet in the lateral geniculate area of the thalamus and from the dorsal raphe nuclei as well [204], [205] and [284]. Importantly for the present model, the SCN has connections with constituents of the fronto-limbic circuitry involved in mood and mood disorders such as the amygdala, the bed nucleus of the stria terminalis, shown to be involved in depression in rats [232], [233] and [279], the paraventricular nuclei of the hypothalamus and the thalamus as well as with the frontal cortical areas [317], [323] and [324]. There is evidence that the SCN may be involved in affective behavior; bilateral lesions of the structure prevent depression in forced swimming [306]. Blockade of prokineticin 2 production, a humoral output molecule for the SCN, has antidepressant effect in forced swimming in mice [170].
Recent evidence from clinical practice [32] and [100] as well as animal research [124] indicates that the wavelength of light exposure may also be critical in that blue but not red light is effective in treating depression in humans and in providing antidepressant effect in forced swimming in rats.
3. Audition <Hearing>
Along with vision, audition is probably the most effective sensory modality in regulating mood. Humans assess emotions and affect not just through visual feedback but also by auditory perception [31], [66] and [67]. There is bidirectional relationship between hearing and mood and its disorders; several types of auditory stimulation can induce positive or negative mood, while depression can alter auditory perception.
3.1. Auditory stimulation modulates mood
Music has been one of the most universally employed means of regulating mood [1] and [134]. Several studies indicate that auditory stimulation in the form of musical compositions, tonal stimulation or simply noise can modulate mood. Listening to different types of music can induce [238], [291] and [310] or alleviate [31], [120], [238], [291], [310] and [312] depressive or anxious mood depending on the nature of the composition [4], [55] and [95].
Forms of auditory stimulation other than musical pieces can also alter mood. For instance, binaural auditory beats induced by presenting two tones differing slightly in frequency to the two ears modulate mood and psychomotor performance [158]. A more negative mood was induced when the perceptual binaural auditory beats were created so as to entrain EEG activity in the theta/delta range (1.5–4 Hz) compared to beta wave range (16–24 Hz). Attesting to the wide range of auditory stimulation that can affect mood, musically enhanced birdsong delivered early in the morning has been shown to induce positive mood, including a decrease in depression as measured by Beck Depression Inventory (BDI) and Profile of Mood States (POMS) [102].
3.2. Studies with animals
A few studies using animal models reveal the potential negative effects of auditory stimulation on mood. While a study showed that a 1 h exposure to an alarm bell (85 dB) did not significantly affect behavior in a forced swim test conducted either immediately or 24 h after the noise stressor [8], another study reported that exposing rats to white noise in a novel environment for 30 min significantly reduced struggling behavior in a swim test compared to controls [330]. Audiogenic stress in the form of high intensity tonal stimulation induced depression-like symptoms in rats in forced swim tests [41]. In contrast to studies finding depressive effect of aversive auditory stimulation in animals, there does not seem to be any published studies in animals reporting positive affective effect of auditory stimulation paralleling findings with humans.
3.3. Hearing impairment affects mood
Further evidence for the contribution of hearing to mood is provided by several studies showing a correlation between problems of hearing and depressive symptoms. Psychiatric comorbidity including depression is more likely in patients with inner ear disorders than in healthy controls or those affected by physical diseases [25]. There is an increased rate of depressive symptoms in people with hearing impairment or total deafness [164], [165], [182], [300] and [351]. Tinnitus, a hearing disorder involving perception of sound without an external source, is significantly correlated with depression [40], [302] and [303].
Hearing deprivation can result in depression in rats as well; depriving rats of auditory stimulation by means of chemically induced temporary deafness results in depression as measured by increased immobility in forced swim tests compared to controls. Remarkably, this effect was abolished when hearing was restored [72].
3.4. Depression affects auditory processing
The bidirectional nature of the interaction between depression and auditory processing is shown by several reports that depression can alter various aspects of auditory processing. For example, depression can impair cortical habituation to auditory stimulation (in the form of vowels or sine tones) in the left auditory cortex as measured by magnetocephalography [311]. Similarly, an fMRI study indicated altered habituation to 800-Hz sine tones in the auditory cortex due to depression [197]. An earlier study reported that compared to controls depression did not alter the pleasure attributed to a subjectively pleasant pure tone, but significantly reduced the repeatability of pleasure derived from the given tone [56].
Hearing loss and auditory perceptual asymmetry are also reported in depression [38] and [183]. Using pure tone and click audiometry, Yovell et al. [340] reported that MDD patients displayed bilateral hearing loss and auditory asymmetry in the form of poorer hearing in the left ear. ECT treatment helped alleviate the asymmetry but not the bilateral hearing loss, regardless of the clinical outcome. Similarly, several studies report auditory perceptual asymmetry in depressed patients with abnormal dichotic listening related to a reduction in left-ear (right-hemisphere) advantage for nonverbal tests reliably observed in non-depressed subjects [39], [127] and [222]. Bruder et al. [37] found that antidepressant (fluoxetine) treatment does not alleviate the abnormal dichotic asymmetry in major depressive patients, suggesting a possible stable characteristic in these patients. The study showed that patients who responded to fluoxetine treatment had greater pre-treatment right-ear advantage for dichotic words and reduced left-ear advantage for complex tones compared to nonresponders.
3.5. Neuroanatomy
Research with animals indicate that audiogenic stress has consequences similar to those observed in other forms of stress both in the way such stimulation affects the HPA axis [30], [42] and [122] and related structures that are involved in stress and mood, particularly the amygdala and the hippocampus [35], [64], [198], [211], [255], [287] and [288].
In humans, a neuroimaging study employing classical music compositions, evaluated by the subjects to be highly pleasurable, activated mainly brain structures involved in emotion and reward, namely the orbitofrontal and ventral medial prefrontal cortices, the amygdala and the ventral striatum, when activity due to neutral music was subtracted [26]. On the other hand, significant rCBF increases in the lateral amygdala/claustrum region have been reported in response to a series of aversive auditory stimuli presented for 60 s (including scraping and scratching sounds at 85–90 dB) compared to white noise [347].
In an elaborate study using six variations of a novel melody with increasing dissonance/unpleasantness (decreasing consonance/pleasantness), Blood et al. [27] found that regional blood flow (rCBF) in specific brain areas covaried with the subjective pleasantness of the stimuli. Specifically increasing dissonance correlated with activity in the right parahippocampal gyrus and precuneus area, whereas increasing consonance correlated with activation of the orbitofrontal, subcallosal cingulate and the frontal pole cortex. The parahippocampal gyrus is reciprocally connected to the amygdala [298] and the frontal and cingulate regions figure prominently in the affective neural circuitry involved in mood regulation. Importantly for the present hypothesis, a case study of a patient with amusia indicated a dissociation between perceptual and emotional analysis of music such that a correct affective assessment (sad or happy) was possible despite failure to recognize or identify the melody [231]. A related study investigated blood oxygenation level dependent (BOLD) signal responses to happy, sad or neutral pieces of music. All three types of auditory stimuli activated the insula and the association auditory cortex. Additionally, happy pieces activated the ventral and dorsal striatum, anterior cingulate, parahippocampal gyrus areas while sad pieces evoked responding in the hippocampus and the amygdala [202].
4. Olfaction <Smell>
It is a well-known fact that both humans and animals are affected by odors adversely or pleasantly depending on the nature and intensity of the odorant, with strong modulatory effect on emotions and mood [200], [201], [272], [273], [282] and [316].
4.1. Modulation of mood by means of odorants <Aromatherapy>
Several studies with nonclinical subjects indicate that some odorants can have either a positive or negative effect on mood. For instance, brief exposure to lavender odor alleviates mood in young subjects regardless of whether their pre-test Beck Depression Inventory scores indicated a depressed or non-depressed state [103]. In a pre–post test comparison without a control odorant, lavender fragrance induced positive mood and lowered anxiety levels in healthy adult subjects as measured by POMS and State Trait Anxiety Inventory (STAI), respectively [92]. Lehner et al. [163] showed that compared to the control condition of no odorant, women exposed to ambient orange odor had more positive mood and lower levels of state anxiety. A recent study reported that mood as measured by PANAS (Positive and Negative Affect Schedule) mood scale was significantly alleviated compared to baseline levels by lemon oil exposure in male and female subjects but not by lavender oil or control treatment [137]. Vanillin, a pleasant odor, elevates mood, while hydrogen sulfide, an unpleasant one, affects mood in an opposite manner [325], paralleling the results of an earlier study comparing vanillin and dimethyl sulfide [142].
4.2. Studies with animals
Odorants, specifically essential oils, have ameliorative effect on depression in animal models as well. A 1 h exposure to vaporized lemon odor and citral repeated at 24 h and 1 h before the second of two consecutive swim tests separated by 24 h had antidepressant effect in rats compared to controls as measured by reduced immobility in the second swim test. Importantly, lemon odor potentiated the effect of the antidepressant imipramine in the forced swim test. This synergism is important from the viewpoint of the proposed model, indicating that sensory stimulation as well as antidepressant treatment may be activating the same central mechanisms in their ameliorative effect on depression and mood [145]. Similar results were obtained in mice exposed to vaporized odorants for 90 min before a forced swim test: lemon oil but not ethanol, lavender or rose had an antidepressant effect as measured by reduced immobility. The study also indicated that the serotonergic system may be involved in the antidepressant effect since the lemon oil odorant significantly increased serotonin turnover in the prefrontal cortex and the striatum [144].
4.3. Olfaction in depression
The central circuitry for olfactory processing has many common links with the affective fronto-limbic circuitry for mood as shown below. An altered olfactory perception therefore can be expected in depression. While olfactory sensitivity is reported to be unaltered in non-seasonal depression [7] and [189] and either unaltered [220] or increased [242] in seasonal depression, studies on the emotional nature of odorants have reported modified olfactory function in depressed patients in early stages of sensory processing as well as in emotional evaluative stages [226], [227], [228] and [229]. A study showed that MDD patients who were comparable in the olfactory task before antidepressant treatment significantly increased their sensitivity to iso-amyl acetate 6 weeks after treatment onset, while depressive symptoms improved 3 weeks earlier [108]. In most cases, however, a reduced rather than increased sensitivity to pure olfactory (rather than trigeminal) cues has been reported in depressed patients with relatively severe symptoms [175], [225], [227] and [281]. Pause et al. [225] reported a reduction in olfactory sensitivity in depression proportional to the severity of the symptoms which was subsequently alleviated by antidepressant treatment. Similarly, a negative correlation was observed between sensitivity to odorants and the level of depression in healthy subjects as measured by BDI [240].
A psychophysiological study [227] confirmed impaired olfactory processing due to depression. These authors assessed central processing of olfactory cues by means of chemosensory event-related potentials (CSERPs) in MDD patients and healthy controls to a pleasant and an unpleasant odor. Results indicated impairment in the early level of olfactory stimulus processing demonstrated by reduced frontal P2 and P3-1 peak amplitudes in depression. Similarly, CSERPs showed impairment in early aspect of olfactory processing when transient learned helplessness was induced in healthy subjects by means of an insolvable social discrimination test [159].
4.4. Neuroanatomy
The olfactory bulbs are neurally connected with brain structures related to mood and depression, particularly the fronto-limbic structures involved in mood modulation by sensory input: the orbitofrontal cortex, insular cortex, the cingulate cortex, the hypothalamus, the amygdala and, importantly for the sensorimotor integration in depression proposed here, the caudate nucleus [9], [106], [209], [251], [252] and [270]. Functional neuroimaging studies in humans indicate that odors activate the piriform cortex, as well as the orbitofrontal cortex, the cingulate cortex, insular cortex, the amygdala and the hypothalamus [277], [345], [346], [349] and [350].
A study by Rolls et al. [253] suggests that there may be a hedonic map for smell in brain regions such as the orbitofrontal cortex. These authors showed that pleasant but not unpleasant odors activated a medial region of the rostral orbitofrontal cortex, with a positive correlation between subjective pleasantness ratings of the six odors used in the experiment and the level of cortical activation. In contrast, a correlation between the subjective unpleasantness ratings of the odors was correlated with activation in the left and more lateral orbitofrontal cortex. While the anterior cingulate cortex was activated by both types of odors, activity in more rostral aspect of the cingulate cortex correlated with pleasantness of the odors.
4.5. Olfactory bulbectomy
A functional model that can contribute to the understanding of the relationship between the sense of smell and depressed mood is the olfactory bulbectomized (OB) rat model of depression. Bilateral olfactory bulbectomy in rats induces several behavioral and physiological disturbances paralleling those observed in humans with affective illnesses that can be attenuated by chronic (but not acute) treatment with antidepressants [34] and [132]. Olfactory bulbectomy leads to hyper-activity and increased corticosterone levels, and produces irritability and anhedonia [133] and [177]. Anatomical and neurochemical studies in rats point to widespread bulbectomy-induced alterations in the functions of fronto-limbic circuitry, including the frontal cortex, the hippocampus and particularly the amygdala, which receives dense innervation from the olfactory bulbs [126] and [295].
5. Taste
There is evidence for a bidirectional interaction between food and mood states; the picture, however, is less clear than that for other modalities, possibly due to complications in specifying the nature, intensity and duration of gustatory stimuli as well as the timing of their impact on the individual via the digestive system.
5.1. Mood induction by tastants
There is evidence that a hedonic response to sweet tastes and carbohydrates may alter mood both in nonclinical subjects and depressed patients. Supporting this view are observations that in dysphoric mood and depression there can be strong craving for food [117] and [118], sweet tasting foods [54] and [128] and chocolate [178]. A hedonic response to sweet tastes and carbohydrates may alter mood both in nonclinical [128] and depressed patients with pronounced craving for sweets and carbohydrates [54] and [256]. Studies with nonclinical subjects indicate that eating a chocolate bar or an apple can elevate mood compared to eating nothing in healthy female subjects [179]. Similarly, palatable but not unpalatable chocolate briefly improves negative mood induced by a film clip [180].
Food preferences in individuals may correlate with the impact of sweet food on mood. A study by Lieberman et al. [173] found that some obese subjects craved carbohydrates in snacking while others had no special preference for carbohydrates over protein-rich snacks. Mood assessment showed that snacking had opposite effects on the two groups in that carbohydrate cravers felt less and noncravers more depressed after eating than before.
5.2. Effects of depression on taste and food intake
Depressed people often complain of dulled sensations of taste which last throughout the depressive episode [301]. While this may be partially due to the fact that many antidepressants modulate monoamine function that can distort or attenuate taste perception [2], [271], [274] and [275], a study nevertheless reports that depressed patients report an ‘unpleasant taste’ regardless of previous antidepressant treatment [199].
Severely depressed patients have decreased sensitivity especially to sweet tastes which normalizes on recovery. Steiner et al. [301] reported that taste recognition thresholds for sucrose but not for other solutions were significantly elevated in depressed patients, and that the amount of threshold elevation was positively correlated with the severity of depression. Amsterdam et al. [7] extended this finding to suprathreshold concentrations of sucrose by demonstrating that intensity ratings grow more slowly with concentration in depressed patients than in normal controls.
There are also reports that suggest a correlation between depressive symptoms and sensitivity to tastants or food preferences. A study with a nonclinical population indicated that thresholds for detecting quinine in solutions were correlated with scores on Beck Depression Inventory such that subjects with higher quinine thresholds also had higher BDI scores [68]. Interestingly, administration of a 5-HT reuptake inhibitor to healthy individuals significantly reduced taste thresholds for quinine as well as for sucrose [112].
5.3. Altered appetite in depression
There is strong evidence that depression modulates appetite; according to one study that investigated appetite change in depression, 66% of the subjects showed decreased and 14% increased appetite while there was no change in 20% of the cases [230].
5.4. Increased appetite and taste
Up to 15% of depressed patients show increased appetite and weight gain [196] and [230]; increased appetite is strongly correlated with sweet cravings [110]. Depression can alter serotonergic levels and craving for carbohydrates and, therefore, sweets may be a compensatory mechanism initiated by lowered central 5-HT levels [91]. In seasonal affective disorder with heightened depressive symptoms in the winter and remission toward the summer, craving for sweets and carbohydrates shows a similar pattern, increasing during the winter and decreasing in the summer [147].
5.5. Decreased appetite and anhedonia
Decreased rather than increased appetite with consequent weight loss is a more common occurrence in depressed patients [230] and in animals [68] and [246]. Anhedonia in the form of reduced appetite and blunted taste reactivity as well as a general loss of interest in food is considered one of the three major components of depression [276].
Anhedonia in the form of reduced intake of a sweet solution induced by chronic mild stress has also been proposed as a model of depression in animals [333], [334] and [341]. Reduction in sucrose or saccharin intake and preferences in rats can be induced by unpredictable mild stress and reversed by a tricyclic antidepressant [335]. A study showed that both a single 5-min exposure to cold-water stress and a 4-week exposure to unpredictable mild stressors resulted in a decreased intake of a saccharin solution in FSL (Flinders Sensitive Line) rats, a putative genetic model of depression [246].
5.6. Neuroanatomy
Compared to other sensory modalities, there is less empirical data on the neuroanatomical substrate of taste perception. In monkeys, studies implicate the dorsal anterior insula and the frontal operculum as the primary and the caudolateral orbitofrontal cortex as the secondary gustatory areas [17], [123], [148], [249] and [254]; a review identified the insula, frontal and parietal opercula as the human cortical gustatory areas [290]. While less is known about the physiology of human than animal central taste mechanisms, neuroimaging studies have begun to provide insight into the impact of tastants on the brain, specifically on the extended limbic circuitry considered in the present model. Functional neuroimaging studies on taste perception indicate activation of not only the primary gustatory cortical areas to tastants but also structures in the fronto-limbic circuitry implicated in mood. A positron emission tomography (PET) study comparing the effects of saline versus pure water in normal volunteers reported differential activation of the insular cortex as well as the anterior cingulate, parahippocampal, temporal and lingual gyri as well as the thalamus and the caudate nucleus [139]. Another PET study demonstrated that an unpleasant tastant, namely quinine, but not a pleasant sugar solution, activated the left amygdala relative to water. Both tastants differentially activated the right posterior orbitofrontal cortex (OFC), while the left anterior OFC responded to the aversive stimulus [343] and [344]. Another study found that both a pleasant sweet taste (1 M glucose) and an unpleasant salt solution (0.1 M NaCl) activated the amygdala bilaterally [217]. Assessment of the pleasantness of gustatory input by the OFC was also demonstrated by an fMRI study that found a significant correlation between OFC activation and decrease in the pleasantness of a liquid food as the subjects became satiated [149].
Both gustatory as well as olfactory cues contribute to the quality of tastants and it can be expected that a convergence of the two modalities will activate central structures implicated in affective processing of sensory inputs. In fact, an fMRI study indicated co-processing of taste-olfactory cues in the caudal orbitofrontal cortex, the insula, anterior cingulate cortex as well as the amygdala [65].
Review
Sensorimotor modulation of mood and depression: An integrative review
R. Canbeyli Corresponding Author Contact Information, E-mail The Corresponding Author
Psychobiology Laboratory, Department of Psychology, Bogazici University, 34342 Bebek, Istanbul, Turkey
Received 18 August 2009; revised 29 October 2009; Accepted 2 November 2009. Available online 11 November 2009.
Abstract
Several lines of research on mood disorders reveal that depression involves a dysfunction in an affective fronto-limbic circuitry that involves the prefrontal cortices, the cingulate cortex, several limbic structures including the amygdala and the hippocampus, lower brainstem structures and the basal ganglia. In dealing with both depressive symptoms as well as their manifestation in the brain, clinical as well as basic research has emphasized mainly a top-down or central approach in elucidating the etiology of depression and its therapy. The present integrative review emphasizes the bottom-up or peripheral view of evaluating the impact of stimulation via sensory modalities or the motor system on the same circuitry in effecting mood regulation and possibly causing mood disorders, specifically depression. The paper shows that there is now a considerable accumulation of data from clinical observations as well as research with animal models to suggest that hypo- or hyper-activation of the sensory or the motor systems by manipulating visual, auditory, olfactory or gustatory inputs as well as physical exercise can have modulatory effects on mood and depressive symptoms. Moreover, depression in turn affects sensorimotor processing, resulting in an interaction that may further contribute to the aggravation of depressive symptoms. The paper also cites evidence that activation of the affective circuitry by central manipulations such as by means of deep brain stimulation has similar modulatory effects as peripheral stimulation on mood and depression. Finally, it is proposed that systematic investigations using animal models of depression on the impact of uni- or multisensory manipulation as well as of physical exercise may provide new insights into the etiology and treatment of depression in humans.
Keywords: Depression; Mood; Mood disorders; Peripheral sensory activation; Exercise
Article Outline
1. Introduction
2. Vision
2.1. Light alleviates depressive symptoms
2.2. Alleviation of depressive symptoms in animals
2.3. Inadequate light photoreception aggravates depressive symptoms
2.4. Animal studies
2.5. Emotion and mood can affect visual processing
2.6. Neuroanatomy
2.7. Regulation of mood by visual activity in the fronto-limbic circuitry
2.8. Mood regulatory action of light via the suprachiasmatic nucleus
3. Audition
3.1. Auditory stimulation modulates mood
3.2. Studies with animals
3.3. Hearing impairment affects mood
3.4. Depression affects auditory processing
3.5. Neuroanatomy
4. Olfaction <Aromatherapy>
4.1. Modulation of mood by means of odorants
4.2. Studies with animals
4.3. Olfaction in depression
4.4. Neuroanatomy
4.5. Olfactory bulbectomy
5. Taste
5.1. Mood induction by tastants
5.2. Effects of depression on taste and food intake
5.3. Altered appetite in depression
5.4. Increased appetite and taste
5.5. Decreased appetite and anhedonia
5.6. Neuroanatomy
6. Multisensory stimulation
6.1. Peripheral stimulation
6.2. Stress and depression
6.3. Enriched environment
6.4. Central stimulation
6.5. Transcranial magnetic stimulation (TMS)
6.6. Vagal nerve stimulation (VNS)
6.7. Deep brain stimulation (DBS)
7. Exercise
7.1. Exercise improves mood and alleviates depression
7.2. Lack of exercise or exercise withdrawal induces depressive symptoms
7.3. Animal research
7.4. Neuroanatomy
8. Conclusion
8.1. Sensorimotor regulation of mood acts via its over- or under-activation of a central circuitry
8.2. The bidirectional nature of sensorimotor manipulation of mood may provide new insight into depression
8.3. Crossmodal interaction of sensory inputs on mood regulation may be relevant to mood regulation
8.4. Neuroanatomical considerations
8.5. Cellular considerations
Acknowledgments
References
1. Introduction
Major depression, second only to hypertension as the most frequent illness not only has severe consequences for the affected individuals but also to their families and social circles. Despite a long history of clinical investigations and basic research into the etiology of depression, the neurobiological basis of mood disorders is still unknown and successful outcome of treatment either in the form of psychotherapy and/or pharmacothreapeutics is far from certain [136] and [263]. Research on the neurobiology of mood disorders has uncovered no biomarker pathognomonic to depression; in fact evidence strongly suggests that depression is not due to a focal dysfunction in the brain. Neuroimaging studies of depressed patients as well as research with animal models indicate that depression is likely to arise out of a dysfunction in a circuitry comprising several cortical and subcortical structures. Observations based on neuroimaging studies with depressed patients and lesion-induced or secondary depression due to neuropathologies such as Parkinson's disease as well as studies with animal models have begun to converge on a fronto-limbic circuitry as the potential basis for mood regulation and mood disorders [78], [113], [114], [143], [214], [218], [236], [237], [245] and [307]. The main cortical components of this circuitry include the ventromedial and dorsal prefrontal cortices (VMPFC and DPFC) and the anterior cingulate cortex. Together with these cortical structures, and forming an extensive fronto-limbic circuit are subcortical structures that include the amygdala, the hippocampus, the hypothalamus as well as the raphe nuclei, nucleus accumbens and the locus coeruleus. The basal ganglia form an integral part of the circuitry not only because of their involvement in psychomotor impairments in depression, but also because of their direct contribution to an affective-motor system [5], [6], [292] and [293].
In light of the circuitry depicted above, and in accordance with the view that depression is a dysfunction of a system rather than of a focal central mechanism, clinical applications as well as basic research on the etiology and treatment of depression have emphasized the complex affective and cognitive situations that figure in mood disorders. With this ‘top-down’ approach, precipitation of depression in humans and induction of depressive symptoms in animals have been considered in light of and attributed to configurations of stressful situations [193], [194], [283] and [305]. Thus, depression is considered an outcome of central dysfunction induced by such stressful situations, undoubtedly affected by genetic factors [84] and [213], reflected as depressive symptoms in behavior and cognitive functions [186] and [187]. While this approach has produced many new insights into and possible therapeutic tools for mood disorders, the present integrative review aims to show that a diametrically opposite approach is not only possible, but may also be productive in understanding and possibly treating mood disorders. Specifically, the present paper takes as its starting point the extensive accumulation of data on the impact of manipulating sensory modalities and the motor system on both the genesis and therapy of mood disorders. As will be shown in subsequent sections, a wealth of findings culled from clinical data as well as studies with animals suggest that, depending on the intensity and duration, modulation of sensory input or motor activity can have ameliorative or debilitating effects on mood. The present paper, therefore, starts from a bottom-up approach by considering the contribution of unisensory as well as multisensory inputs to mood disorders. Additionally, the model considers the motor/behavioral side of the underlying mechanism not as a mere output conduit for psychomotor symptoms of depression but as possible cause of and potential basis for treatment for such symptoms.
The proposed bottom-up approach to mood is predicated on the hypothesis that hypo- or hyper-activation of a sensory modality or of the motor system can have a modulatory effect on mood in general and depression in particular. The proposed approach also emphasizes the bidirectional nature of sensorimotor activation and depression, considering not only sensorimotor manipulation of mood, but also the effect of depression on sensorimotor processing. It will be argued that the interactive nature of sensorimotor activation and depression is not only a major complication in the clinical setting, but the realization of its implications may actually open new avenues of animal research on depression and eventual insights into both the etiology and treatment of mood disorders. While there is as yet no clear assessment of the relationship between mood regulation and the level of sensorimotor activation, the present proposal will suggest that there is adequate empirical evidence to warrant a systematic approach to the psychophysical evaluation of the impact of the level of sensorimotor activation on mood regulation and depression.
The bottom-up approach to mood regulation proposed in the present paper proposes that the activation of an ‘affective-motor’ circuitry by sensory stimulation or physical activity is capable of modulating mood. In support of this view, it will be shown in one of the following sections that central as well as peripheral activation of the proposed circuitry does modulate mood and affect depressive symptoms. The present review does not aim to provide a detailed account of the involvement of the extended neural circuitry in the proposed model. There are several reviews that address this issue directly and show, among many techniques, that neuroimaging studies indicate changed neuroanatomical characteristics of the structures involved in the affective circuitry (for a recent review see for example Ref. [176]). However, each section below will point to studies that demonstrate a relationship between the specific sensorimotor system discussed and the neuroanatomical structures in the proposed circuitry that figure in mood modulation in that modality.
The present paper is not meant to be an exhaustive assessment of sensorimotor manipulations relevant to mood and its disorders, but as an integrative evaluation aimed to emphasize the relevance of such manipulations to mood when systematic/parametric evaluation of such manipulations have been made (for this reason, for instance, the sense of touch is not explicitly discussed because of the rather informal/unsystematic manner in which tactile stimulation has been applied as an adjunct treatment). Likewise, discussion is limited to unipolar depression.
2.0 Vision
In addition to its all important role in providing sight through the visual system, light can affect mood in both humans and animals. The modulatory nature of the relationship between photoreception and mood is shown by the fact that photic stimulation can alleviate, whereas impaired vision or inadequate light reception can aggravate depressive symptoms.
2.1. Light alleviates depressive symptoms
Regulation of mood by light is a universally known phenomenon experienced by many around the globe as a result of short or long days in the winter and summer, respectively. In contrast to the other sensory modalities to be discussed in the present paper, affective impact of photic stimulation has been investigated systematically so that light therapy is now a widely used therapeutic tool for the treatment of depression. Starting in 1980s light therapy in the form of exposure to a bright source of light for approximately an hour either in the morning or early evening has been used successfully to alleviate symptoms of both seasonal [10], [81], [105], [167], [168], [221], [257], [258], [326] and [327] and non-seasonal depression [88], [151] and [181]. There is also evidence that exposure to artificial bright light can have ameliorative effect on mood in nonclinical populations. A study conducted in Finland in the winter indicated that repeated exposure to at least 1 h of light in office workers reduced depressive symptoms in healthy subjects with or without winter-related mood symptoms [224]. Interestingly, another Finnish study with nonclinical subjects indicated that bright light exposure (2500–4000 lx at eye level) paired with exercise over an 8-week period significantly reduced depressive symptoms compared to exercise in normal ambient illumination (400–600 lx) [166].
2.2. Alleviation of depressive symptoms in animals
There are a few studies that indicate that exposure to light, particularly in the form of long photoperiods has ameliorative effect on depressive symptoms in animals. Prolonged exposure of male rats to a long photoperiod (15 h:9 h light/dark cycle) for 30 days has an antidepressant effect as measured by reduced immobility in a forced swim test [203]. The ameliorative effect of light treatment was comparable to 30 days of treatment with the antidepressants clomipramine or desipramine. There are also indications that a much shorter exposure to light exposure can have protective effect on depression in the rat. Yilmaz et al. [338] showed that a single 12-h light exposure in female rats maintained normally on a 12 h:12 h L/D cycle has antidepressant effect as measured by behavioral despair. Subsequent studies have shown that even shorter light pulses of 30 min [278] or 10 min duration [124] delivered late in the dark phase of an L/D cycle have ameliorative effect in behavioral despair, an animal model of depression [241].
2.3. Inadequate light photoreception aggravates depressive symptoms
Just as enhanced photic stimulation is conducive to good mood and can alleviate depressive symptoms, low levels of illumination have the opposite effect of leading to negative mood and aggravating depressive symptoms [87], [125] and [152] and depression is more prevalent in areas receiving little sunlight throughout the year [29]. Given these findings, it can be expected that people with visual impairment that reduces the level of experienced illumination should show more depressive symptoms. This is in fact confirmed by clinical studies that report significant rates of depressive symptoms in patients with ophthalmic dysfunctions that compromise photoreception and/or reduce visual acuity [260], [261] and [285]. Within the framework of the sensorimotor model presented here, it should be emphasized that another factor contributing to the increased depressive symptoms in people with compromised photoreception, particularly in older subjects, is reduced motility due to impaired vision. In fact, a study that explicitly investigated activity levels along with photoreception in age-related macular degeneration (AMD) indicated that activity loss due to vision problems was a mediating factor in the relationship between visual acuity and depressive symptoms [259].
2.4. Animal studies
A 3-week exposure to a short photoperiod (5 h light/9 h dark cycle) induces depression-like behavior in sand rats compared to controls maintained on a 12 h light/12 h dark cycle [85]. Compared to animals maintained on a long-day regimen (16 h light/8 h dark days), similar depression-like results have been reported in male but not female Siberian hamsters that were exposed for 3 weeks to short days (8 h light/16 h dark cycle) starting at the time of weaning [244]. Another study with Siberian hamsters indicated that perinatal exposure to short days (8 h light/day) for 20–23 days after birth leads in adulthood to more depressive-like behavior in forced swimming compared to exposure to long days (16 h light/8 h dark cycle) [247].
2.5. Emotion and mood can affect visual processing
Just as visual input affects mood, emotions and depression can in turn modulate visual processing. Recent neuroimaging studies indicate that emotions can affect central visual processing even at the earliest stages of processing [3], [14] and [195]. A few studies have also begun to show that ‘emotional vision’ is particularly affected by depression. An fMRI study with depressed (MDD) patients, for example, showed reduced activity in the right anterior cingulate cortex and the left insula to positive pictures and in the right insula and hippocampus to negative pictures [162]. There was also a significant correlation between severity of depression and activity induced by negative pictures in the left amygdala, the left insula and bilateral inferior orbitofrontal cortex. Another study comparing event-related potentials (ERPs) to colors and emotional slides in MDD patients and healthy subjects found reduced responses in late aspects of visual processing in MDD patients (decreased amplitudes of P3 and pSW) which recovered after pharmacotherapy [227].
2.6. Neuroanatomy
Light can affect mood via two major pathways. In its non-visual function, light cannot only affect the proposed limbic circuitry via the amygdala (by thalamic relay) and frontal cortical pathways, it can also modulate mood by its action on the suprachiasmatic nucleus (SCN) that regulates many biological rhythms on a daily and yearly basis.
2.7. Regulation of mood by visual activity in the fronto-limbic circuitry
Visual input can modulate mood by a cascade of events that starts with the visual cortices, continuing in the dorsal and the ventral visual projections involving the parietal and inferotemporal cortices, respectively. The ventral stream in turn projects to the fronto-limbic structures involved in affective processing, namely the orbitofrontal cortex, the ventromedial prefrontal cortex, the cingulate gyrus, the hippocampus and the amygdala. These structures, particularly the amygdala, can induce depressive symptoms via their to connections with the hypothalamus, the basal ganglia and brainstem structures including the locus coeruleus and the raphe nuclei [48], [49], [150], [161], [172], [250] and [314]. The amygdala figures prominently in affective processing of visual input because of its connections with many cortical structures involved in visual processing including the striate cortex, the temporal cortex, and the frontal lobes as well as subcortical structures implicated in depression [160] and [235]. The fact that fear in a face can be detected by people with damage to the striate cortex (those with ‘blind sight’) [67] but not by those with damage to the amygdala [319] attests not only to the critical role of the amygdala in emotional processing, but also indicates that emotional visual processing can occur without conscious awareness and without the detailed processing that is achieved by the visual cortical areas.
2.8. Mood regulatory action of light via the suprachiasmatic nucleus
In addition to the circuitry depicted above, light input can modulate mood via its effect on the SCN, the master oscillator in the hypothalamus that regulates many circadian as well as circannual biorhythms. Briefly, light acts a zeitgeber to entrain SCN so that the large number of biological rhythms are eurythmic throughout the day. Moreover, the relative lengths of the day (light) and night (darkness) portions of the day show regular changes throughout the year, thereby providing the basis for the biological clock to regulate rhythms on a yearly basis [206], [207] and [315]. While social cues and other factors can also entrain the SCN, light is the primary input to the biological clock. Information about light is sent to the SCN via a direct retinohypothalamic tract, or indirectly via projections from the intergeniculate leaflet in the lateral geniculate area of the thalamus and from the dorsal raphe nuclei as well [204], [205] and [284]. Importantly for the present model, the SCN has connections with constituents of the fronto-limbic circuitry involved in mood and mood disorders such as the amygdala, the bed nucleus of the stria terminalis, shown to be involved in depression in rats [232], [233] and [279], the paraventricular nuclei of the hypothalamus and the thalamus as well as with the frontal cortical areas [317], [323] and [324]. There is evidence that the SCN may be involved in affective behavior; bilateral lesions of the structure prevent depression in forced swimming [306]. Blockade of prokineticin 2 production, a humoral output molecule for the SCN, has antidepressant effect in forced swimming in mice [170].
Recent evidence from clinical practice [32] and [100] as well as animal research [124] indicates that the wavelength of light exposure may also be critical in that blue but not red light is effective in treating depression in humans and in providing antidepressant effect in forced swimming in rats.
3. Audition <Hearing>
Along with vision, audition is probably the most effective sensory modality in regulating mood. Humans assess emotions and affect not just through visual feedback but also by auditory perception [31], [66] and [67]. There is bidirectional relationship between hearing and mood and its disorders; several types of auditory stimulation can induce positive or negative mood, while depression can alter auditory perception.
3.1. Auditory stimulation modulates mood
Music has been one of the most universally employed means of regulating mood [1] and [134]. Several studies indicate that auditory stimulation in the form of musical compositions, tonal stimulation or simply noise can modulate mood. Listening to different types of music can induce [238], [291] and [310] or alleviate [31], [120], [238], [291], [310] and [312] depressive or anxious mood depending on the nature of the composition [4], [55] and [95].
Forms of auditory stimulation other than musical pieces can also alter mood. For instance, binaural auditory beats induced by presenting two tones differing slightly in frequency to the two ears modulate mood and psychomotor performance [158]. A more negative mood was induced when the perceptual binaural auditory beats were created so as to entrain EEG activity in the theta/delta range (1.5–4 Hz) compared to beta wave range (16–24 Hz). Attesting to the wide range of auditory stimulation that can affect mood, musically enhanced birdsong delivered early in the morning has been shown to induce positive mood, including a decrease in depression as measured by Beck Depression Inventory (BDI) and Profile of Mood States (POMS) [102].
3.2. Studies with animals
A few studies using animal models reveal the potential negative effects of auditory stimulation on mood. While a study showed that a 1 h exposure to an alarm bell (85 dB) did not significantly affect behavior in a forced swim test conducted either immediately or 24 h after the noise stressor [8], another study reported that exposing rats to white noise in a novel environment for 30 min significantly reduced struggling behavior in a swim test compared to controls [330]. Audiogenic stress in the form of high intensity tonal stimulation induced depression-like symptoms in rats in forced swim tests [41]. In contrast to studies finding depressive effect of aversive auditory stimulation in animals, there does not seem to be any published studies in animals reporting positive affective effect of auditory stimulation paralleling findings with humans.
3.3. Hearing impairment affects mood
Further evidence for the contribution of hearing to mood is provided by several studies showing a correlation between problems of hearing and depressive symptoms. Psychiatric comorbidity including depression is more likely in patients with inner ear disorders than in healthy controls or those affected by physical diseases [25]. There is an increased rate of depressive symptoms in people with hearing impairment or total deafness [164], [165], [182], [300] and [351]. Tinnitus, a hearing disorder involving perception of sound without an external source, is significantly correlated with depression [40], [302] and [303].
Hearing deprivation can result in depression in rats as well; depriving rats of auditory stimulation by means of chemically induced temporary deafness results in depression as measured by increased immobility in forced swim tests compared to controls. Remarkably, this effect was abolished when hearing was restored [72].
3.4. Depression affects auditory processing
The bidirectional nature of the interaction between depression and auditory processing is shown by several reports that depression can alter various aspects of auditory processing. For example, depression can impair cortical habituation to auditory stimulation (in the form of vowels or sine tones) in the left auditory cortex as measured by magnetocephalography [311]. Similarly, an fMRI study indicated altered habituation to 800-Hz sine tones in the auditory cortex due to depression [197]. An earlier study reported that compared to controls depression did not alter the pleasure attributed to a subjectively pleasant pure tone, but significantly reduced the repeatability of pleasure derived from the given tone [56].
Hearing loss and auditory perceptual asymmetry are also reported in depression [38] and [183]. Using pure tone and click audiometry, Yovell et al. [340] reported that MDD patients displayed bilateral hearing loss and auditory asymmetry in the form of poorer hearing in the left ear. ECT treatment helped alleviate the asymmetry but not the bilateral hearing loss, regardless of the clinical outcome. Similarly, several studies report auditory perceptual asymmetry in depressed patients with abnormal dichotic listening related to a reduction in left-ear (right-hemisphere) advantage for nonverbal tests reliably observed in non-depressed subjects [39], [127] and [222]. Bruder et al. [37] found that antidepressant (fluoxetine) treatment does not alleviate the abnormal dichotic asymmetry in major depressive patients, suggesting a possible stable characteristic in these patients. The study showed that patients who responded to fluoxetine treatment had greater pre-treatment right-ear advantage for dichotic words and reduced left-ear advantage for complex tones compared to nonresponders.
3.5. Neuroanatomy
Research with animals indicate that audiogenic stress has consequences similar to those observed in other forms of stress both in the way such stimulation affects the HPA axis [30], [42] and [122] and related structures that are involved in stress and mood, particularly the amygdala and the hippocampus [35], [64], [198], [211], [255], [287] and [288].
In humans, a neuroimaging study employing classical music compositions, evaluated by the subjects to be highly pleasurable, activated mainly brain structures involved in emotion and reward, namely the orbitofrontal and ventral medial prefrontal cortices, the amygdala and the ventral striatum, when activity due to neutral music was subtracted [26]. On the other hand, significant rCBF increases in the lateral amygdala/claustrum region have been reported in response to a series of aversive auditory stimuli presented for 60 s (including scraping and scratching sounds at 85–90 dB) compared to white noise [347].
In an elaborate study using six variations of a novel melody with increasing dissonance/unpleasantness (decreasing consonance/pleasantness), Blood et al. [27] found that regional blood flow (rCBF) in specific brain areas covaried with the subjective pleasantness of the stimuli. Specifically increasing dissonance correlated with activity in the right parahippocampal gyrus and precuneus area, whereas increasing consonance correlated with activation of the orbitofrontal, subcallosal cingulate and the frontal pole cortex. The parahippocampal gyrus is reciprocally connected to the amygdala [298] and the frontal and cingulate regions figure prominently in the affective neural circuitry involved in mood regulation. Importantly for the present hypothesis, a case study of a patient with amusia indicated a dissociation between perceptual and emotional analysis of music such that a correct affective assessment (sad or happy) was possible despite failure to recognize or identify the melody [231]. A related study investigated blood oxygenation level dependent (BOLD) signal responses to happy, sad or neutral pieces of music. All three types of auditory stimuli activated the insula and the association auditory cortex. Additionally, happy pieces activated the ventral and dorsal striatum, anterior cingulate, parahippocampal gyrus areas while sad pieces evoked responding in the hippocampus and the amygdala [202].
4. Olfaction <Smell>
It is a well-known fact that both humans and animals are affected by odors adversely or pleasantly depending on the nature and intensity of the odorant, with strong modulatory effect on emotions and mood [200], [201], [272], [273], [282] and [316].
4.1. Modulation of mood by means of odorants <Aromatherapy>
Several studies with nonclinical subjects indicate that some odorants can have either a positive or negative effect on mood. For instance, brief exposure to lavender odor alleviates mood in young subjects regardless of whether their pre-test Beck Depression Inventory scores indicated a depressed or non-depressed state [103]. In a pre–post test comparison without a control odorant, lavender fragrance induced positive mood and lowered anxiety levels in healthy adult subjects as measured by POMS and State Trait Anxiety Inventory (STAI), respectively [92]. Lehner et al. [163] showed that compared to the control condition of no odorant, women exposed to ambient orange odor had more positive mood and lower levels of state anxiety. A recent study reported that mood as measured by PANAS (Positive and Negative Affect Schedule) mood scale was significantly alleviated compared to baseline levels by lemon oil exposure in male and female subjects but not by lavender oil or control treatment [137]. Vanillin, a pleasant odor, elevates mood, while hydrogen sulfide, an unpleasant one, affects mood in an opposite manner [325], paralleling the results of an earlier study comparing vanillin and dimethyl sulfide [142].
4.2. Studies with animals
Odorants, specifically essential oils, have ameliorative effect on depression in animal models as well. A 1 h exposure to vaporized lemon odor and citral repeated at 24 h and 1 h before the second of two consecutive swim tests separated by 24 h had antidepressant effect in rats compared to controls as measured by reduced immobility in the second swim test. Importantly, lemon odor potentiated the effect of the antidepressant imipramine in the forced swim test. This synergism is important from the viewpoint of the proposed model, indicating that sensory stimulation as well as antidepressant treatment may be activating the same central mechanisms in their ameliorative effect on depression and mood [145]. Similar results were obtained in mice exposed to vaporized odorants for 90 min before a forced swim test: lemon oil but not ethanol, lavender or rose had an antidepressant effect as measured by reduced immobility. The study also indicated that the serotonergic system may be involved in the antidepressant effect since the lemon oil odorant significantly increased serotonin turnover in the prefrontal cortex and the striatum [144].
4.3. Olfaction in depression
The central circuitry for olfactory processing has many common links with the affective fronto-limbic circuitry for mood as shown below. An altered olfactory perception therefore can be expected in depression. While olfactory sensitivity is reported to be unaltered in non-seasonal depression [7] and [189] and either unaltered [220] or increased [242] in seasonal depression, studies on the emotional nature of odorants have reported modified olfactory function in depressed patients in early stages of sensory processing as well as in emotional evaluative stages [226], [227], [228] and [229]. A study showed that MDD patients who were comparable in the olfactory task before antidepressant treatment significantly increased their sensitivity to iso-amyl acetate 6 weeks after treatment onset, while depressive symptoms improved 3 weeks earlier [108]. In most cases, however, a reduced rather than increased sensitivity to pure olfactory (rather than trigeminal) cues has been reported in depressed patients with relatively severe symptoms [175], [225], [227] and [281]. Pause et al. [225] reported a reduction in olfactory sensitivity in depression proportional to the severity of the symptoms which was subsequently alleviated by antidepressant treatment. Similarly, a negative correlation was observed between sensitivity to odorants and the level of depression in healthy subjects as measured by BDI [240].
A psychophysiological study [227] confirmed impaired olfactory processing due to depression. These authors assessed central processing of olfactory cues by means of chemosensory event-related potentials (CSERPs) in MDD patients and healthy controls to a pleasant and an unpleasant odor. Results indicated impairment in the early level of olfactory stimulus processing demonstrated by reduced frontal P2 and P3-1 peak amplitudes in depression. Similarly, CSERPs showed impairment in early aspect of olfactory processing when transient learned helplessness was induced in healthy subjects by means of an insolvable social discrimination test [159].
4.4. Neuroanatomy
The olfactory bulbs are neurally connected with brain structures related to mood and depression, particularly the fronto-limbic structures involved in mood modulation by sensory input: the orbitofrontal cortex, insular cortex, the cingulate cortex, the hypothalamus, the amygdala and, importantly for the sensorimotor integration in depression proposed here, the caudate nucleus [9], [106], [209], [251], [252] and [270]. Functional neuroimaging studies in humans indicate that odors activate the piriform cortex, as well as the orbitofrontal cortex, the cingulate cortex, insular cortex, the amygdala and the hypothalamus [277], [345], [346], [349] and [350].
A study by Rolls et al. [253] suggests that there may be a hedonic map for smell in brain regions such as the orbitofrontal cortex. These authors showed that pleasant but not unpleasant odors activated a medial region of the rostral orbitofrontal cortex, with a positive correlation between subjective pleasantness ratings of the six odors used in the experiment and the level of cortical activation. In contrast, a correlation between the subjective unpleasantness ratings of the odors was correlated with activation in the left and more lateral orbitofrontal cortex. While the anterior cingulate cortex was activated by both types of odors, activity in more rostral aspect of the cingulate cortex correlated with pleasantness of the odors.
4.5. Olfactory bulbectomy
A functional model that can contribute to the understanding of the relationship between the sense of smell and depressed mood is the olfactory bulbectomized (OB) rat model of depression. Bilateral olfactory bulbectomy in rats induces several behavioral and physiological disturbances paralleling those observed in humans with affective illnesses that can be attenuated by chronic (but not acute) treatment with antidepressants [34] and [132]. Olfactory bulbectomy leads to hyper-activity and increased corticosterone levels, and produces irritability and anhedonia [133] and [177]. Anatomical and neurochemical studies in rats point to widespread bulbectomy-induced alterations in the functions of fronto-limbic circuitry, including the frontal cortex, the hippocampus and particularly the amygdala, which receives dense innervation from the olfactory bulbs [126] and [295].
5. Taste
There is evidence for a bidirectional interaction between food and mood states; the picture, however, is less clear than that for other modalities, possibly due to complications in specifying the nature, intensity and duration of gustatory stimuli as well as the timing of their impact on the individual via the digestive system.
5.1. Mood induction by tastants
There is evidence that a hedonic response to sweet tastes and carbohydrates may alter mood both in nonclinical subjects and depressed patients. Supporting this view are observations that in dysphoric mood and depression there can be strong craving for food [117] and [118], sweet tasting foods [54] and [128] and chocolate [178]. A hedonic response to sweet tastes and carbohydrates may alter mood both in nonclinical [128] and depressed patients with pronounced craving for sweets and carbohydrates [54] and [256]. Studies with nonclinical subjects indicate that eating a chocolate bar or an apple can elevate mood compared to eating nothing in healthy female subjects [179]. Similarly, palatable but not unpalatable chocolate briefly improves negative mood induced by a film clip [180].
Food preferences in individuals may correlate with the impact of sweet food on mood. A study by Lieberman et al. [173] found that some obese subjects craved carbohydrates in snacking while others had no special preference for carbohydrates over protein-rich snacks. Mood assessment showed that snacking had opposite effects on the two groups in that carbohydrate cravers felt less and noncravers more depressed after eating than before.
5.2. Effects of depression on taste and food intake
Depressed people often complain of dulled sensations of taste which last throughout the depressive episode [301]. While this may be partially due to the fact that many antidepressants modulate monoamine function that can distort or attenuate taste perception [2], [271], [274] and [275], a study nevertheless reports that depressed patients report an ‘unpleasant taste’ regardless of previous antidepressant treatment [199].
Severely depressed patients have decreased sensitivity especially to sweet tastes which normalizes on recovery. Steiner et al. [301] reported that taste recognition thresholds for sucrose but not for other solutions were significantly elevated in depressed patients, and that the amount of threshold elevation was positively correlated with the severity of depression. Amsterdam et al. [7] extended this finding to suprathreshold concentrations of sucrose by demonstrating that intensity ratings grow more slowly with concentration in depressed patients than in normal controls.
There are also reports that suggest a correlation between depressive symptoms and sensitivity to tastants or food preferences. A study with a nonclinical population indicated that thresholds for detecting quinine in solutions were correlated with scores on Beck Depression Inventory such that subjects with higher quinine thresholds also had higher BDI scores [68]. Interestingly, administration of a 5-HT reuptake inhibitor to healthy individuals significantly reduced taste thresholds for quinine as well as for sucrose [112].
5.3. Altered appetite in depression
There is strong evidence that depression modulates appetite; according to one study that investigated appetite change in depression, 66% of the subjects showed decreased and 14% increased appetite while there was no change in 20% of the cases [230].
5.4. Increased appetite and taste
Up to 15% of depressed patients show increased appetite and weight gain [196] and [230]; increased appetite is strongly correlated with sweet cravings [110]. Depression can alter serotonergic levels and craving for carbohydrates and, therefore, sweets may be a compensatory mechanism initiated by lowered central 5-HT levels [91]. In seasonal affective disorder with heightened depressive symptoms in the winter and remission toward the summer, craving for sweets and carbohydrates shows a similar pattern, increasing during the winter and decreasing in the summer [147].
5.5. Decreased appetite and anhedonia
Decreased rather than increased appetite with consequent weight loss is a more common occurrence in depressed patients [230] and in animals [68] and [246]. Anhedonia in the form of reduced appetite and blunted taste reactivity as well as a general loss of interest in food is considered one of the three major components of depression [276].
Anhedonia in the form of reduced intake of a sweet solution induced by chronic mild stress has also been proposed as a model of depression in animals [333], [334] and [341]. Reduction in sucrose or saccharin intake and preferences in rats can be induced by unpredictable mild stress and reversed by a tricyclic antidepressant [335]. A study showed that both a single 5-min exposure to cold-water stress and a 4-week exposure to unpredictable mild stressors resulted in a decreased intake of a saccharin solution in FSL (Flinders Sensitive Line) rats, a putative genetic model of depression [246].
5.6. Neuroanatomy
Compared to other sensory modalities, there is less empirical data on the neuroanatomical substrate of taste perception. In monkeys, studies implicate the dorsal anterior insula and the frontal operculum as the primary and the caudolateral orbitofrontal cortex as the secondary gustatory areas [17], [123], [148], [249] and [254]; a review identified the insula, frontal and parietal opercula as the human cortical gustatory areas [290]. While less is known about the physiology of human than animal central taste mechanisms, neuroimaging studies have begun to provide insight into the impact of tastants on the brain, specifically on the extended limbic circuitry considered in the present model. Functional neuroimaging studies on taste perception indicate activation of not only the primary gustatory cortical areas to tastants but also structures in the fronto-limbic circuitry implicated in mood. A positron emission tomography (PET) study comparing the effects of saline versus pure water in normal volunteers reported differential activation of the insular cortex as well as the anterior cingulate, parahippocampal, temporal and lingual gyri as well as the thalamus and the caudate nucleus [139]. Another PET study demonstrated that an unpleasant tastant, namely quinine, but not a pleasant sugar solution, activated the left amygdala relative to water. Both tastants differentially activated the right posterior orbitofrontal cortex (OFC), while the left anterior OFC responded to the aversive stimulus [343] and [344]. Another study found that both a pleasant sweet taste (1 M glucose) and an unpleasant salt solution (0.1 M NaCl) activated the amygdala bilaterally [217]. Assessment of the pleasantness of gustatory input by the OFC was also demonstrated by an fMRI study that found a significant correlation between OFC activation and decrease in the pleasantness of a liquid food as the subjects became satiated [149].
Both gustatory as well as olfactory cues contribute to the quality of tastants and it can be expected that a convergence of the two modalities will activate central structures implicated in affective processing of sensory inputs. In fact, an fMRI study indicated co-processing of taste-olfactory cues in the caudal orbitofrontal cortex, the insula, anterior cingulate cortex as well as the amygdala [65].