**3. Functional neuroimaging**

The major limitation of structural neuroimaging techniques is that they are suitable for studying diseases associated with morphologic alterations, such as neurologic conditions. For this reason, they are only partially useful in psychiatric disorders which are characterized by behavioral abnormalities due to neurochemical impairment. In this perspective, PET (Abraham & Feng, 2011) and fMRI represent the gold standard for brain imaging aimed to assess cognitive performance (Glower, 2011). Electroencefalography, Event-Related Potentials and Magnetoencephalography are less specific and, therefore, mostly used to exclude neurological conditions in clinical practice or for research purposes (Cohen & Cuffin, 1983). Medication, drug or alcohol abuse and genetic/epigenetic influence represent major confounding factors (Nakama et al., 2011; Schulte et al., 2010). On the other side, following the biopsychosocial model for psychiatric disorders, functional neuroimaging could help understanding the complex interaction between environmental stressors, genetic risk and precipitating events in the plasticity of neural circuitry and consequently in clinical symptoms.

Studies using phosphorus MRS (31P- MRS) have found phase-specific alterations of phospholipid membranes, high energy phosphates and intracellular brain pH in BD. In particular, a number of investigations reported a reduced intracellular cerebral pH in bipolar subjects which has been associated with the increased levels of lactate observed in some 1H- MRS studies. Both conditions are indicative of a shift from oxidative phosphorylation to glycolysis. There is also a 31P- MRS based-report of decreased levels of

Stork and Renshaw proposed a cohesive model that puts together the majority of MRS findings. They hypothesized that the impaired oxidative phopshorilation, the decreased cellular energy and the altered membrane metabolism could be due to an underlying altered

**Technique Main alterations** 

**1H- MRS** N- acetyl aspartate Reduced levels

**31P- MRS** Phosphocreatine Reduced levels

The major limitation of structural neuroimaging techniques is that they are suitable for studying diseases associated with morphologic alterations, such as neurologic conditions. For this reason, they are only partially useful in psychiatric disorders which are characterized by behavioral abnormalities due to neurochemical impairment. In this perspective, PET (Abraham & Feng, 2011) and fMRI represent the gold standard for brain imaging aimed to assess cognitive performance (Glower, 2011). Electroencefalography, Event-Related Potentials and Magnetoencephalography are less specific and, therefore, mostly used to exclude neurological conditions in clinical practice or for research purposes (Cohen & Cuffin, 1983). Medication, drug or alcohol abuse and genetic/epigenetic influence represent major confounding factors (Nakama et al., 2011; Schulte et al., 2010). On the other side, following the biopsychosocial model for psychiatric disorders, functional neuroimaging could help understanding the complex interaction between environmental stressors, genetic risk and precipitating events in the plasticity of neural circuitry and

Choline Increased levels Glutamate, GABA and Glutamine Increased levels Myo-inositol Increased levels in

Lactate Increased levels

Phosphomonoesters Reduced levels Intracellular brain pH Reduced levels

mania and euthymia and reduced levels in bipolar depression

phosphocreatine and of phosphomonoesters in BD (Kato et al., 1995).

mitochondrial metabolism in BD (Stork & Renshaw, 2005). Main MRS findings in BD are synthetized in Table 2.

Table 2. Main MRS findings in BD.

**3. Functional neuroimaging** 

consequently in clinical symptoms.

Functional neuroimaging attempts to explain psychiatric disorders by means of degenerative or developmental model of illness and/or in terms of hypometabolism.

In fact, elevated activity of the hippocampus or of the ventral prefrontal cortex as well as dorsolateral prefrontal cortex hypofunction are recurrent themes in literature (Savitz & Drevets, 2009).

#### **3.1 Positron Emission Tomography (PET) and Single Photon Emission Computed Tomography (SPECT)**

PET imaging is a direct measure of a radioactive decay due to cerebral metabolism of a radioactive substance or radionuclide. Different body tissues are characterized by different consumption rates of radionuclides (Ter-Pogossian et al., 1975; Vyas et al., 2011). Radionuclides used in clinical practice are usually major compounds of biologic molecules (18-Fluorine in the form of 18-Fluorodeoxyglucose or FDG for measuring glucose metabolism, 15-Oxygen for measuring blood flow, 11-Carbon or 13-Nitrogen common in diagnostic PET procedures). The nuclide is introduced in the patient and the radioactive decay is measured (Phelps et al., 1975): in particular, the positron emitted by nuclides has a collision with electrons producing a gamma photon which is measured by the PET camera (Roncali & Cherry, 2011). PET can measure both blood flow and glucose metabolism, often used as surrogate measures of neuronal synaptic activity. A first line comparison is between the neuroligand uptake in target regions and reference area while a more complex analysis can compare blood flow or glucose uptake in the same subject in different states, i.e. while resting or during a cognitive performance. Both ways provide useful data for research and clinical analysis; anyway, a major limitation is the use of a radioactive nuclide. Specifically, targeted PET radioligands are used to investigate neurotransmitter systems (Weisel, 1989). Cerebral PET has its major use in neurological disease: excluding primary or secondary oncologic lesions, evaluation of dementia, confirming epilepsy or assessing the state in cerebrovascular disease (Cavalcanti et al., 2011; Mazzuca et al., 2011; Person et al., 2010; Quigley et al., 2010; Salas and Gonzales, 2011).

SPECT works capturing orbiting electrons without a positron-electron collision, but by means of an emission of a single photon by the SPECT nuclide. Main nuclides used in SPECT are 123-Iodine, 33-Technetium or 133-Xenon. Single photons are selected with the use of multiple collimators.

PET and SPECT studies in depressive disorders have shown that blood rate and flow are increased both in BD and in unipolar depression in the frontal lobes during depressive episodes. However, they are increased during mania in the dorsal cingulate cortex, striatal regions, and the nucleus accumbens, as well as in limbic structures of the temporal lobes and reduced in dorsolateral prefrontal cortex, possibly reflecting its loss of modulatory control over limbic structures (Gonul et al., 2009).

With respect to neurotransmitters, serotonin (5-HT) transporter was found to have an increased density in the thalamus (Laje et al., 2010), dorsal cingulate cortex, medial prefrontal cortex and insula of depressed BD patients. 5-HT has been implicated in mania as well: in particular, individuals with current mania had significantly lower 5-HT2 receptor binding potential in frontal, temporal, parietal and occipital cortical regions, with more prominent changes in the right cortical regions compared to controls (Yatham et al., 2010, 2002a, 2002b). With regards to 5-HT1A receptor, bipolar depressed patients were found to show higher 5-HT1A in raphe nuclei and forebrain (Sullivan et al., 2009). An interesting use of PET consists of assessing the role of serotonin in major depressive episodes comparing

Neuroimaging Data in Bipolar Disorder: An Updated View 271

or elevated left orbitofrontal activity. Right dorsolateral prefrontal cortical activity was found to be reduced, while the increased left prefrontal activity seems to be a state marker of

The few studies with fMRI on manic patients report an increased activity of the amygdala, insular cortex and subcortical areas in response to negative emotional stimuli. Ventral striatal activity was found to be elevated at rest and during motor tasks. On the other hand, ventral prefrontal activity was found to be attenuated during cognitive performances (Altshuler et al., 2005; Elliott et al., 2004). In addition, bilateral orbitofrontal attenuation has been reported in mania and may represent a trait feature of the disorder as it is also present

Main fMRI alterations

increased on the left

Increased left activity

Increased activity in response to negative stimuli

cognitive performances

Reduced activity during attentional tasks

attentional tasks) or reduced activity (i.e. in response to fearful stimuli, during working memory tasks)

performance or working memory tasks

to fearful stimuli

to fearful stimuli; Significantly increased activity in response to reward stimuli

Orbitofrontal cortex Activity reduced bilaterally or

Prefrontal cortex Reduced right activity;

Ventral prefrontal cortex Reduced activity during

Dorsolateral prefrontal cortex Increased (i.e. during

Orbitofrontal cortex Reduced activity bilaterally

Subcortical areas Increased activity during

Amygdala Increased activity in response

Striatum Increased activity in response

structures involved

Insula Subcortical areas

Anterior cingulate gyrus

Findings on euthymic bipolar patients are more consistent and have pointed out reduced activity in dorsal, ventral prefrontal cortical regions and dorsal regions of the anterior cingulate gyrus during performance of attentional tasks. Dorsolateral prefrontal cortical activity was found to be, on the contrary, increased. Other studies have reported reduced dorsolateral prefrontal cortex activity in euthymic individuals during working memory and verbal encoding tasks (Deckersbach et al, 2006; Monks et al., 2004). Increases in activity

bipolar depression (Altshuler et al., 2008).

during bipolar depression (Altshuler et al., 2008).

**Bipolar depression** 

BD Phase Central nervous system

**Mania** Amygdala

**Euthymia** Ventral prefrontal cortex

Table 3. Main fMRI findings in BD.

BD vs unipolar depression. In fact, both unipolar and bipolar depression were associated with elevated 5-HT transporter binding in the insula, thalamus and striatum, but showed distinct abnormalities in the brainstem (Cannon et al., 2007).

With respect to dopamine, D1 receptor binding potentials were found to be reduced in frontal cortex, even though striatal D2 receptor density was normal in all phases of nonpsychotic BD (Bauer, 2003; Suhara et al., 1992).

In synthesis, PET and SPECT studies have shown in BD a loss of modulatory control of the cortex over limbic structures, reflected by specific phase-dependent modifications of blood rate and flow. Alterations of neurotransmitters involved in the pathogenesis of BD have also been reported, particularly with respect to serotonin transporter, serotonin receptor density and dopamine receptor density.

#### **3.2 Functional Magnetic Resonance Imaging (fMRI)**

FMRI is the most used technique in brain mapping and in psychiatric research due to its non-invasive technology, wide availability, high spatial and temporal resolution and the lack of ionizing radiation that allows the clinician to repeat functional exams over time as well as in different phases of illness. FMRI, in fact, is suitable for studying bipolar patients' performances on the same cognitive tasks during depressed, manic or euthymic phases. It can also compare brain activity during symptom exacerbation as well as over periods of remission.

One limit of fMRI is that it gives limited information on subcortical structures. Spatial resolution remains anyway highly relevant for the study of psychiatric diseases, given the clear correlation between cortical dysfunction and many psychiatric symptoms. Another limit consists of the increased variance of the results obtained with this technique in psychiatric patients (Dougherty et al., 2004).

FMRI measures changes in blood flow in areas of the central nervous system (Konarsky et al., 2007). The hemodynamic response reflects neural activity in the brain or spinal cord as neurons have no reserve for oxygen or glucose and they need to rapidly increase blood flow when necessary. A Blood-oxygen-level dependent (BOLD) signal is measured by fMRI. From a physiological perspective, hemoglobin is diamagnetic when oxygenated (oxyhemoglobin) and paramagnetic when deoxygenated (deoxyhemoglobin) producing different signals that are higher when coming from activated areas. Actually, an increase in cerebral blood flow produces changes in oxygen consumption resulting in increased BOLD signals (Bandettini, 2003).

Studies with fMRI in bipolar patients showed various alterations of the activity in different regions of the cortico-limbic pathways responsible for emotional regulation: amygdala, thalamus, striatum, portions of the prefrontal cortex and anterior cingulated cortex. Studies, however, were limited by the small samples size and by the possible interference of the medication. The increased activation of amygdala, striatum and thalamus were the most constant findings among the different studies (Cerullo et al., 2009).

Increased amygdala and subcortical activity to emotional stimuli, in particular negative stimuli, as well as reduced activity of the prefrontal cortical regions during cognitive performances are common to all phases of BD, suggesting that they may be trait features of the disease (Phillips & Vieta, 2007). Other additional frontal and temporal regions were found to be activated, maybe as a compensatory mechanism (Townsed et al., 2010).

FMRI studies in bipolar patients also suggest the presence of phase-dependent abnormalities. In fact, bipolar depression is associated with attenuated bilateral orbitofrontal

BD vs unipolar depression. In fact, both unipolar and bipolar depression were associated with elevated 5-HT transporter binding in the insula, thalamus and striatum, but showed

With respect to dopamine, D1 receptor binding potentials were found to be reduced in frontal cortex, even though striatal D2 receptor density was normal in all phases of non-

In synthesis, PET and SPECT studies have shown in BD a loss of modulatory control of the cortex over limbic structures, reflected by specific phase-dependent modifications of blood rate and flow. Alterations of neurotransmitters involved in the pathogenesis of BD have also been reported, particularly with respect to serotonin transporter, serotonin receptor density

FMRI is the most used technique in brain mapping and in psychiatric research due to its non-invasive technology, wide availability, high spatial and temporal resolution and the lack of ionizing radiation that allows the clinician to repeat functional exams over time as well as in different phases of illness. FMRI, in fact, is suitable for studying bipolar patients' performances on the same cognitive tasks during depressed, manic or euthymic phases. It can also compare brain activity during symptom exacerbation as well as over periods of

One limit of fMRI is that it gives limited information on subcortical structures. Spatial resolution remains anyway highly relevant for the study of psychiatric diseases, given the clear correlation between cortical dysfunction and many psychiatric symptoms. Another limit consists of the increased variance of the results obtained with this technique in

FMRI measures changes in blood flow in areas of the central nervous system (Konarsky et al., 2007). The hemodynamic response reflects neural activity in the brain or spinal cord as neurons have no reserve for oxygen or glucose and they need to rapidly increase blood flow when necessary. A Blood-oxygen-level dependent (BOLD) signal is measured by fMRI. From a physiological perspective, hemoglobin is diamagnetic when oxygenated (oxyhemoglobin) and paramagnetic when deoxygenated (deoxyhemoglobin) producing different signals that are higher when coming from activated areas. Actually, an increase in cerebral blood flow produces changes in oxygen consumption resulting in increased BOLD

Studies with fMRI in bipolar patients showed various alterations of the activity in different regions of the cortico-limbic pathways responsible for emotional regulation: amygdala, thalamus, striatum, portions of the prefrontal cortex and anterior cingulated cortex. Studies, however, were limited by the small samples size and by the possible interference of the medication. The increased activation of amygdala, striatum and thalamus were the most

Increased amygdala and subcortical activity to emotional stimuli, in particular negative stimuli, as well as reduced activity of the prefrontal cortical regions during cognitive performances are common to all phases of BD, suggesting that they may be trait features of the disease (Phillips & Vieta, 2007). Other additional frontal and temporal regions were

FMRI studies in bipolar patients also suggest the presence of phase-dependent abnormalities. In fact, bipolar depression is associated with attenuated bilateral orbitofrontal

found to be activated, maybe as a compensatory mechanism (Townsed et al., 2010).

constant findings among the different studies (Cerullo et al., 2009).

distinct abnormalities in the brainstem (Cannon et al., 2007).

psychotic BD (Bauer, 2003; Suhara et al., 1992).

**3.2 Functional Magnetic Resonance Imaging (fMRI)** 

psychiatric patients (Dougherty et al., 2004).

signals (Bandettini, 2003).

and dopamine receptor density.

remission.

or elevated left orbitofrontal activity. Right dorsolateral prefrontal cortical activity was found to be reduced, while the increased left prefrontal activity seems to be a state marker of bipolar depression (Altshuler et al., 2008).

The few studies with fMRI on manic patients report an increased activity of the amygdala, insular cortex and subcortical areas in response to negative emotional stimuli. Ventral striatal activity was found to be elevated at rest and during motor tasks. On the other hand, ventral prefrontal activity was found to be attenuated during cognitive performances (Altshuler et al., 2005; Elliott et al., 2004). In addition, bilateral orbitofrontal attenuation has been reported in mania and may represent a trait feature of the disorder as it is also present during bipolar depression (Altshuler et al., 2008).


Table 3. Main fMRI findings in BD.

Findings on euthymic bipolar patients are more consistent and have pointed out reduced activity in dorsal, ventral prefrontal cortical regions and dorsal regions of the anterior cingulate gyrus during performance of attentional tasks. Dorsolateral prefrontal cortical activity was found to be, on the contrary, increased. Other studies have reported reduced dorsolateral prefrontal cortex activity in euthymic individuals during working memory and verbal encoding tasks (Deckersbach et al, 2006; Monks et al., 2004). Increases in activity

Neuroimaging Data in Bipolar Disorder: An Updated View 273

allowing goal and sub-goal structuring. Moreover, it is involved in reward prediction. On the other hand, the temporal pole seems to be activated in basic semantic processes with person-emotion linkages associated with narrative. BD patients have a deficit of performance on visuospatial and constructional praxis which suggests an atypical localization of cognitive functions. This atypical localization and the hyperconnectivity between specific regions could be responsible for the enhanced creativity and writing ability

Recently, abnormalities in perigenual anterior cingulate cortex-amygdala functional connectivity during emotional processing have been found in BD (Wang et al., 2009). Similar findings have been reported even in children and adolescents with BD, concluding that in these subjects significant white matter tract alterations were present in regions involved in emotional, behavioural and cognitive regulation. In addition, these results suggest that alterations in white matter are present early in the course of disease in familial BD (Barnea-Goraly et al., 2009; Kavafaris et al., 2009). An impaired fiber density in anterior corona radiata (as detected with a decreased fractional anisotropy) was detected in BD in pediatric age and in Attention Deficit and Hyperactivity Disorder suggesting a possible link

DTI studies can allow to detect a possible overlap between BD and schizophrenia. In fact, reduced integrity of the anterior limb of the internal capsule, uncinate fasciculus and anterior thalamic radiation regions is common to both schizophrenia and BD suggesting an overlap in white matter pathology, possibly relating to risk factors common to both

Concerning antidepressants and mood stabilizers, these compounds seem to have neuroprotective effects and are not likely to explain white matter abnormalities, even though minor effects cannot be excluded (Bruno et al., 2008). Anyway, microstructural abnormality in

In synthesis, DTI provides information on functional connectivity between regions of the central nervous system. DTI studies on bipolar probands showed a reduced gray matter in areas such as putamen, caudate nucleus, nucleus acumbens, insula and orbitofrontal cortex. As concerns white matter, connections between orbitofrontal cortex, temporal, parietal corteces and the frontosubcortical circuits were found to be altered during mania and also over euthymia, as possible traits of BD. DTI findings have interesting implications on the association between BD and creativity. The hyperconnectivity between specific regions and the atypical localization of cognitive functions seem to be correlated to the enhanced creativity and writing ability of BD subjects. On the other hand, the atypical localization of cognitive functions could underlie the visuoconstructional praxis deficit present in BD.

Since the introduction of CT, researchers focused their efforts in elucidating the connection between psychiatric diseases and the presence of structural cerebral alterations through neuroimaging. CT pioneered this research without providing, however, a complete answer. Actually, a growing body of evidence has been accumulated in literature as newer techniques such as MRI and functional imaging (i.e., SPECT, PET, fMRI) have been introduced revealing much about the biological underpinnings of neuropsychiatric disorders. Neuroimaging research in BD has already produced several data documenting the involvement of different cortical and subcortical regions in different phases of the

the white matter has been associated with a low remission rate of major depression.

observed in BD probands (McCrea, 2008).

between the two disorders (Pavuluri et al., 2008).

disorders (Sussman et al., 2008).

**4. Conclusions** 

within subcortical regions associated with emotion processing rather than working memory or attention have also been detected in remitted, euthymic individuals with BD during performance of a continuous performance task (Strakowski et al., 2004) and working memory task (Adler et al., 2004). Other studies investigated the response of the activity of these structures to fearful expressions in remitted bipolar patients. Results showed an increased activity in the amygdala and in the striatum and, on the other hand, a reduction of the dorsolateral prefrontal cortex activity (Phillips & Vieta, 2007). Of note, striatal activity in response to potentially rewarding stimuli was found to be significantly elevated. Other emotional stimuli led to decreased dorsolateral prefrontal cortical activity. These two patterns may underlie mood instabilities in euthymic patients, especially in those with comorbidities (Hassel et al., 2008).

In synthesis, fMRI findings in bipolar patients are heterogeneous: they may be present in all phases of BD and/or can be phase-dependent. Among the formers, the most significant data include an increased activity of the amygdala and of the subcortical areas to negative stimuli and a reduced activity of the prefrontal cortex during cognitive tasks. Bipolar depression has been associated with modifications of the activity of the orbitofrontal and prefrontal cortex. In mania, specific alterations include an increased activity of the striatum at rest and during motor tasks and a reduction of the prefrontal cortex activity during cognitive performances. There are several studies on euthymic patients showing modifications of the activity of the prefrontal cortex during attentional or working memory tasks. Structures implicated in the emotional processing seem to be involved as well: in fact, modifications of the activity of the amygdala, striatum and dorsolateral prefrontal cortex in response to different emotional stimuli have been reported.

#### **3.3 Diffusion Tensor Imaging (DTI)**

Diffusion tensor imaging (DTI) is an MRI application developed in order to investigate white matter connections between regions of interest. These connections provide information on functional activity between areas of the central nervous system. DTI is particularly useful to detect white matter lesions or dysfunction (Versace et al., 2008).

There are few studies with DTI in BD, most of them based on the promising results from MRI research showing microstructural alterations in white matter in various neocortical areas and in the corpus callosum. In particular, fractional anisotropy, the most sensitive DTI marker which reflects fiber density, axonal diameter and myelination in white matter, was found to be decreased significantly in the ventral part of the corpus callosum in patients with BD (Heller et al., 2011). Other interesting results coming from DTI revealed that gray matter concentration was reduced in BD in the right anterior insula, head of the caudate nucleus, nucleus accumbens, ventral putamen and frontal orbital cortex. Other studies pointed out that BD patients showed abnormalities within white matter tracts connecting the frontal cortex with the temporal and parietal cortices and the fronto-subcortical circuits (Lin et al., 2011). White matter abnormalities seem to persist by the time of remission even after the first manic episodes (Chan et al., 2010), suggesting that disruption of white matter cortical-subcortical networks as well as projection, associative and commissural tracts may be a hallmark of the illness (Heng et al., 2010) involving prefrontal and frontal regions, associative and commissural fibres.

Some recent studies reported that certain variants of BD may be due to an increased functional or effective connectivity between orbitofrontal and temporal pole structures in the dominant hemisphere. The orbitofrontal cortex codifies the value of different stimuli,

within subcortical regions associated with emotion processing rather than working memory or attention have also been detected in remitted, euthymic individuals with BD during performance of a continuous performance task (Strakowski et al., 2004) and working memory task (Adler et al., 2004). Other studies investigated the response of the activity of these structures to fearful expressions in remitted bipolar patients. Results showed an increased activity in the amygdala and in the striatum and, on the other hand, a reduction of the dorsolateral prefrontal cortex activity (Phillips & Vieta, 2007). Of note, striatal activity in response to potentially rewarding stimuli was found to be significantly elevated. Other emotional stimuli led to decreased dorsolateral prefrontal cortical activity. These two patterns may underlie mood instabilities in euthymic patients, especially in those with

In synthesis, fMRI findings in bipolar patients are heterogeneous: they may be present in all phases of BD and/or can be phase-dependent. Among the formers, the most significant data include an increased activity of the amygdala and of the subcortical areas to negative stimuli and a reduced activity of the prefrontal cortex during cognitive tasks. Bipolar depression has been associated with modifications of the activity of the orbitofrontal and prefrontal cortex. In mania, specific alterations include an increased activity of the striatum at rest and during motor tasks and a reduction of the prefrontal cortex activity during cognitive performances. There are several studies on euthymic patients showing modifications of the activity of the prefrontal cortex during attentional or working memory tasks. Structures implicated in the emotional processing seem to be involved as well: in fact, modifications of the activity of the amygdala, striatum and dorsolateral prefrontal cortex in response to

Diffusion tensor imaging (DTI) is an MRI application developed in order to investigate white matter connections between regions of interest. These connections provide information on functional activity between areas of the central nervous system. DTI is

There are few studies with DTI in BD, most of them based on the promising results from MRI research showing microstructural alterations in white matter in various neocortical areas and in the corpus callosum. In particular, fractional anisotropy, the most sensitive DTI marker which reflects fiber density, axonal diameter and myelination in white matter, was found to be decreased significantly in the ventral part of the corpus callosum in patients with BD (Heller et al., 2011). Other interesting results coming from DTI revealed that gray matter concentration was reduced in BD in the right anterior insula, head of the caudate nucleus, nucleus accumbens, ventral putamen and frontal orbital cortex. Other studies pointed out that BD patients showed abnormalities within white matter tracts connecting the frontal cortex with the temporal and parietal cortices and the fronto-subcortical circuits (Lin et al., 2011). White matter abnormalities seem to persist by the time of remission even after the first manic episodes (Chan et al., 2010), suggesting that disruption of white matter cortical-subcortical networks as well as projection, associative and commissural tracts may be a hallmark of the illness (Heng et al., 2010) involving prefrontal and frontal regions,

Some recent studies reported that certain variants of BD may be due to an increased functional or effective connectivity between orbitofrontal and temporal pole structures in the dominant hemisphere. The orbitofrontal cortex codifies the value of different stimuli,

particularly useful to detect white matter lesions or dysfunction (Versace et al., 2008).

comorbidities (Hassel et al., 2008).

different emotional stimuli have been reported.

**3.3 Diffusion Tensor Imaging (DTI)** 

associative and commissural fibres.

allowing goal and sub-goal structuring. Moreover, it is involved in reward prediction. On the other hand, the temporal pole seems to be activated in basic semantic processes with person-emotion linkages associated with narrative. BD patients have a deficit of performance on visuospatial and constructional praxis which suggests an atypical localization of cognitive functions. This atypical localization and the hyperconnectivity between specific regions could be responsible for the enhanced creativity and writing ability observed in BD probands (McCrea, 2008).

Recently, abnormalities in perigenual anterior cingulate cortex-amygdala functional connectivity during emotional processing have been found in BD (Wang et al., 2009). Similar findings have been reported even in children and adolescents with BD, concluding that in these subjects significant white matter tract alterations were present in regions involved in emotional, behavioural and cognitive regulation. In addition, these results suggest that alterations in white matter are present early in the course of disease in familial BD (Barnea-Goraly et al., 2009; Kavafaris et al., 2009). An impaired fiber density in anterior corona radiata (as detected with a decreased fractional anisotropy) was detected in BD in pediatric age and in Attention Deficit and Hyperactivity Disorder suggesting a possible link between the two disorders (Pavuluri et al., 2008).

DTI studies can allow to detect a possible overlap between BD and schizophrenia. In fact, reduced integrity of the anterior limb of the internal capsule, uncinate fasciculus and anterior thalamic radiation regions is common to both schizophrenia and BD suggesting an overlap in white matter pathology, possibly relating to risk factors common to both disorders (Sussman et al., 2008).

Concerning antidepressants and mood stabilizers, these compounds seem to have neuroprotective effects and are not likely to explain white matter abnormalities, even though minor effects cannot be excluded (Bruno et al., 2008). Anyway, microstructural abnormality in the white matter has been associated with a low remission rate of major depression.

In synthesis, DTI provides information on functional connectivity between regions of the central nervous system. DTI studies on bipolar probands showed a reduced gray matter in areas such as putamen, caudate nucleus, nucleus acumbens, insula and orbitofrontal cortex. As concerns white matter, connections between orbitofrontal cortex, temporal, parietal corteces and the frontosubcortical circuits were found to be altered during mania and also over euthymia, as possible traits of BD. DTI findings have interesting implications on the association between BD and creativity. The hyperconnectivity between specific regions and the atypical localization of cognitive functions seem to be correlated to the enhanced creativity and writing ability of BD subjects. On the other hand, the atypical localization of cognitive functions could underlie the visuoconstructional praxis deficit present in BD.
