Modeling panic disorder in rodents.

Panic disorder (PD) is a subtype of anxiety disorder in which the core phenomenon is the spontaneous occurrence of panic attacks. Although studies with laboratory animals have been instrumental for the understanding of its neurobiology and treatment, few review articles have focused on the validity of the currently used animal models for studying this psychopathology. Therefore, the aim of the present paper is to discuss the strengths and limits of these models in terms of face, construct and predictive validity. Based on the hypothesis that panic attacks are related to defensive responses elicited by proximal threat, most animal models measure the escape responses induced by specific stimuli. Some apply electrical or chemical stimulation to brain regions proposed to modulate fear and panic responses, such as the dorsal periaqueductal grey or the medial hypothalamus. Other models focus on the behavioural consequences caused by the exposure of rodents to ultrasound or natural predators. Finally, the elevated T-maze associates a one-way escape response from an open arm with panic attacks. Despite some limitations, animal models are essential for a better understanding of the neurobiology and pharmacology of PD and for discovering more effective treatments.

(1)H magnetic resonance spectroscopy imaging of the hippocampus in patients with panic disorder.

Recent theories of panic disorder propose an extensive involvement of limbic system structures, such as the hippocampus, in the pathophysiology of this condition. Despite this, no prior study has examined exclusively the hippocampal neurochemistry in this disorder. The current study used proton magnetic resonance spectroscopy imaging ((1)H-MRSI) to examine possible abnormalities in the hippocampus in panic disorder patients. Participants comprised 25 panic patients and 18 psychiatrically healthy controls. N-acetylaspartate (NAA, a putative marker of neuronal viability) and choline (Cho, involved in the synthesis and degradation of cell membranes) levels were quantified relative to creatine (Cr, which is thought to be relatively stable among individuals and in different metabolic condition) in both right and left hippocampi. Compared with controls, panic patients demonstrated significantly lower NAA/Cr in the left hippocampus. No other difference was detected. This result is consistent with previous neuroimaging findings of hippocampal alterations in panic and provides the first neurochemical evidence suggestive of involvement of this structure in the disorder. Moreover, lower left hippocampal NAA/Cr in panic disorder may possibly reflect neuronal loss and/or neuronal metabolic dysfunction, and could be related to a deficit in evaluating ambiguous cues.

Neurobiology of panic disorder: from animal models to brain neuroimaging.

Evidence from animal models of anxiety has led to the hypothesis that serotonin enhances inhibitory avoidance (related to anxiety) in the forebrain, but inhibits one-way escape (panic) in the midbrain periaqueductal gray (PAG). Stressing the difference between these emotions, neuroendocrinological results indicate that the hypothalamic-pituitary-adrenal axis is activated by anticipatory anxiety, but not by panic attack nor by electrical stimulation of the rat PAG. Functional neuroimaging has shown activation of the insula and upper brain stem (including PAG), as well as deactivation of the anterior cingulated cortex (ACC) during experimental panic attacks. Voxel-based morphometric analysis of brain magnetic resonance images has shown a grey matter volume increase in the insula and upper brain stem, and a decrease in the ACC of panic patients at rest, as compared to healthy controls. The insula and the ACC detect interoceptive stimuli, which are overestimated by panic patients. It is suggested that these brain areas and the PAG are involved in the pathophysiology of panic disorder.

Regional gray matter abnormalities in panic disorder: a voxel-based morphometry study.

Although abnormalities in brain structures involved in the neurobiology of fear and anxiety have been implicated in the pathophysiology of panic disorder (PD), relatively few studies have made use of voxel-based morphometry (VBM) magnetic resonance imaging (MRI) to determine structural brain abnormalities in PD. We have assessed gray matter volume in 19 PD patients and 20 healthy volunteers using VBM. Images were acquired using a 1.5 T MRI scanner, and were spatially normalized and segmented using optimized VBM. Statistical comparisons were performed using the general linear model. A relative increase in gray matter volume was found in the left insula of PD patients compared with controls. Additional structures showing differential increases were the left superior temporal gyrus, the midbrain, and the pons. A relative gray matter deficit was found in the right anterior cingulate cortex. The insula and anterior cingulate abnormalities may be relevant to the pathophysiology of PD, since these structures participate in the evaluation process that ascribes negative emotional meaning to potentially distressing cognitive and interoceptive sensory information. The abnormal brain stem structures may be involved in the generation of panic attacks.

[Anxiety, panic and the hypothalamic-pituitary-adrenal axis]. / Ansiedade, pânico e o eixo hipotálamo-pituitária-adrenal.

OBJECTIVE: This article focuses on the differential activation of the hypothalamic-pituitary-adrenal axis in generalized anxiety disorder and panic disorder. METHOD: The results of recently reported reviews of the literature are summarized and discussed. RESULTS: The results of experimental studies that assayed adrenocorticotropic hormone, cortisol and prolactin show that real-life panic attacks, as well as those induced by selective panicogenic agents such as lactate and carbon dioxide, do not activate the hypothalamic-pituitary-adrenal axis. Agonists of the cholecystokinin receptor B such as the cholecystokinin-4 peptide and pentagastrin increase stress hormones regardless of the occurrence of a panic attack and, thus, seem to activate the hypothalamic-pituitary-adrenal axis directly. The benzodiazepine antagonist flumazenil does not increase stress hormones, but this agent does not reliably induce panic attacks. Pharmacological agents that increase anxiety in both normal people and panic patients (caffeine, yohimbine, serotonergic agonists) raise stress hormone levels. CONCLUSIONS: In addition to the differences in symptomatology and pharmacological response, generalized anxiety disorder and panic disorder affect stress hormones in distinct ways. While anticipatory anxiety and generalized anxiety disorder activate both the hypothalamic-pituitary-adrenal and the sympathoadrenal axes, panic attack causes major sympathetic activation, but has little effect on the hypothalamic-pituitary-adrenal axis.

Translational approach to the pathophysiology of panic disorder: Focus on serotonin and endogenous opioids.

Panic patients experience recurrent panic attacks. Two main neurochemical hypotheses have been proposed to explain this vulnerability. The first suggests that panic patients have deficient serotonergic inhibition of neurons localized in the dorsal periaqueductal gray matter of the midbrain that organizes defensive reactions to cope with proximal threats as well as of sympathomotor control areas of the rostral ventrolateral medulla that generate neurovegetative symptoms of the panic attack. The second proposes that endogenous opioids buffer panic attacks in normal subjects, and their deficit results in heightened sensitivity to suffocation and separation anxiety in panic patients. Experimental results obtained in rat models of panic indicate that serotonin interacts synergistically with endogenous opioids in the dorsal periaqueductal gray through 5-HT1A and µ-opioid receptors to inhibit proximal defense and, supposedly, panic attacks. These findings allow reconciliation of the serotonergic and opioidergic hypotheses of panic pathophysiology. They also indicate that endogenous opioids are likely to participate in the panicolytic action of antidepressants and suggest that exogenous opioids may be useful for treating panic patients resistant to conventional pharmacotherapy.

Subjective and neurovegetative changes in healthy volunteers and panic patients performing simulated public speaking.

Drug-free symptomatic panic patients, drug-treated nonsymptomatic patients and healthy controls were submitted to simulated public speaking. Subjective anxiety, cognitive impairment and discomfort measured by the visual analog mood scale as well as skin conductance level were higher in symptomatic patients than in controls at the beginning of the experimental session, nonsymptomatic patients lying in between. Subjective sedation, spontaneous fluctuations of skin conductance, heart rate and blood pressure were similar in the three groups. Preparation and performance of speech decreased sedation while increasing anxiety, cognitive impairment, level and fluctuations of skin conductance, heart rate and blood pressure. Anxiety, cognitive impairment and conductance level were less increased in symptomatic patients than in controls. Electrodermal activity, but not cardiovascular measures of sympathetic arousal correlated with anticipatory anxiety. Chronic treatment with serotonin uptake inhibitors attenuated the differences between panic patients and controls, supporting the participation of serotonin in panic disorder.

Anxiety and salivary cortisol in symptomatic and nonsymptomatic panic patients and healthy volunteers performing simulated public speaking.

Anxiety and salivary cortisol were measured in subjects performing simulated public speaking (SPS), a procedure that has been neurobiologically related to panic disorder. The subjects were divided into three groups: 18 symptomatic panic patients, 16 nonsymptomatic, drug-treated panic patients, and 17 healthy controls. In the experimental session, subjective anxiety (Visual Analogue Mood Scale) and the total score of the Bodily Symptom Scale (BSS) were higher in symptomatic patients than in controls, with nonsymptomatic patients in between. Measures of cortisol taken at home showed that the level was higher at 9:00 h than at 23:00 h in every group, indicating a normal circadian regulation of the hypothalamic-pituitary-adrenal (HPA) axis in panic patients. Also in every group, the level of cortisol was high at the beginning of the experimental session and decreased after 70 min. This fall parallels the decrease in anxiety and BSS ratings, and appears to reflect habituation of initial, anticipatory anxiety. Preparation and performance of speech raised anxiety and BSS scores to the initial levels, but failed to increase cortisol measured over 60 min, starting at the end of the speech. Therefore, SPS does not seem to activate the HPA axis, as reported in panic attacks.

Serotonin, the periaqueductal gray and panic.

This article reviews experimental evidence and theoretical constructs that implicate serotonin (5-HT) modulation of defensive behavior within the midbrain periaqueductal gray in panic disorder (PD). Evidence with conflict tests in experimental animals indicates that 5-HT enhances anxiety, whereas results with aversive stimulation of the dorsal periaqueductal gray point to an anxiolytic role of 5-HT. To solve this contradiction, it has been suggested that the emotional states determined by the two types of animal model are different. Conflict tests would generate conditioned anxiety, whereas periaqueductal gray stimulation would produce unconditioned fear, as evoked by proximal threat. Clinically, the former would be related to generalized anxiety while the latter to PD. Thus, 5-HT is supposed to facilitate anxiety, but to inhibit panic. This hypothesis has been tested in the animal model of anxiety and panic named the elevated T-maze, in two procedures of human experimental anxiety applied to healthy volunteers or panic patients, and in CO2-induced panic attacks. Overall, the obtained results have shown that drugs that enhance 5-HT function increase different indexes of anxiety, but decrease indexes of panic. Drugs that impair 5-HT function have the opposite effects. Thus, so far the predictions derived from the above hypothesis have been fulfilled.

On serotonin and experimental anxiety.

BACKGROUND: This review describes the development of a research line on the role of serotonin (5-HT) in experimental anxiety that was initiated in 1969, in the laboratory founded by P.B. Dews, W.H. Morse and R.T. Kelleher at the Harvard Medical School, and has evolved until this date. RESULTS: Initially, it was found that two non-selective 5-HT receptor antagonists released punished responding in pigeons with a magnitude comparable to that of benzodiazepine anxiolytics. This result was one of the key evidences that led to the concept that 5-HT enhanced anxiety by acting both in the forebrain and in the periaqueductal gray matter (PAG). Further evidence supported this hypothesis regarding the forebrain, but results with electrical stimulation and intracerebral drug injection into the PAG indicated that 5-HT inhibited aversive behavior evoked from this area. As a result, it has been suggested that 5-HT has a dual role in the regulation of defense, namely enhancing learned responses to potential or distal threat through actions in the forebrain while inhibiting unconditioned responses to proximal threat by acting on the PAG. The former would be related to generalized anxiety and the latter to panic disorder. To test this hypothesis, a new animal model, named the elevated T-maze, has been designed. It consists of one arm enclosed by walls that is perpendicular to two open arms elevated from the floor. The same rat performs two tasks, namely inhibitory avoidance of the elevated open arms, representing conditioned anxiety and one-way escape from one of the open arms, representative of unconditioned fear. CONCLUSION: The differential effects of drugs acting on 5-HT observed in the two tasks of the ETM generally support the hypothesis under scrutiny.

Anxiolytic-like effects of median raphe nucleus lesion in the elevated T-maze.

The cell bodies of 5-HT containing neurons that innervate the limbic forebrain are mainly found in the dorsal raphe nucleus and in the median raphe nucleus (MRN). To assess the role of the median raphe nucleus in anxiety, rats bearing either electrolytic or 5-HT-selective neurotoxic lesion of the MRN were tested in the elevated T-maze. This apparatus consists of two opposed open arms perpendicular to one enclosed arm. Two tasks are performed in succession by the same rat in one experimental session, namely inhibitory avoidance of the open arm, taken as a measure of conditioned anxiety and one-way escape from the open arm, considered as a measure of unconditioned fear. The test was performed 7 days after the electrolytic lesion (3 mA, 10s) or 14 days after the neurotoxic lesion (5,7-DHT, 8 microg/1 microl). The results showed that while the electrolytic lesion impaired both inhibitory avoidance and one-way escape, the neurotoxic lesion impaired only inhibitory avoidance. Therefore, serotonergic pathways originating in the MRN seem to participate in the modulation of conditioned anxiety but not unconditioned fear. Other neurotransmitter systems that either originate in or pass through the MRN may regulate unconditioned fear.

Executive and modulatory neural circuits of defensive reactions: implications for panic disorder.

The present review covers two independent approaches, a neuroanatomical and a pharmacological (focused on serotonergic transmission), which converge in highlighting the critical role of the hypothalamus and midbrain periaqueductal gray matter in the generation of panic attacks and in the mechanism of action of current antipanic medication. Accordingly, innate and learned fear responses to different threats (i.e., predator, aggressive members of the same species, interoceptive threats and painful stimuli) are processed by independent circuits involving corticolimbic regions (the amygdala, the hippocampus and the prefrontal and insular cortices) and downstream hypothalamic and brainstem circuits. As for the drug treatment, animal models of panic indicate that the drugs currently used for treating panic disorder should work by enhancing 5-HT inhibition of neural systems that command proximal defense in both the dorsal periaqueductal gray and in the medial hypothalamus. For the anticipatory anxiety, the reviewed evidence points to corticolimbic structures, such as the amygdala, the septo-hippocampus and the prefrontal cortex, as its main neural substrate, modulated by stimulation of 5-HT2C and 5-HT1A receptors.

Serotonin in anxiety and panic: contributions of the elevated T-maze.

The elevated T-maze (ETM) was developed to test the hypothesis that serotonin (5-HT) plays an opposing role in the regulation of defensive behaviors associated with anxiety and panic. This test allows the measurement in the same rat of inhibitory avoidance acquisition, related to generalized anxiety disorder, and of one-way escape, associated with panic disorder. The evidence so far reported with the ETM supports the above hypothesis and indicates that: (1) whereas 5-HT neurons located at the dorsal raphe nucleus are involved in the regulation of both inhibitory avoidance and escape, those of the median raphe nucleus are primarily implicated in the former task; (2) facilitation of 5-HT1A- and 5-HT2A-mediated neurotransmission in the dorsal periaqueductal gray (dPAG) is likely to mediate the panicolytic drug action; (3) stimulation of 5-HT2C receptors in the basolateral amygdala increases anxiety and is implicated in the anxiogenesis caused by short-term administration of antidepressant drugs, and (4) 5-HT1A and the µ-opioid receptors work together in the dPAG to modulate escape or panic attacks. These last results point to the possible benefits of adjunctive opioid therapy for panic patients resistant to antidepressants that act on 5-HT neurotransmission.

Interaction between µ-opioid and 5-HT1A receptors in the regulation of panic-related defensive responses in the rat dorsal periaqueductal grey.

A wealth of evidence indicates that the activation of 5-HT1A and 5-HT2A receptors in the dorsal periaqueductal grey matter (dPAG) inhibits escape, a panic-related defensive behaviour. Results that were previously obtained with the elevated T-maze test of anxiety/panic suggest that 5-HT1A and µ-opioid receptors in this midbrain area work together to regulate this response. To investigate the generality of this finding, we assessed whether the same cooperative mechanism is engaged when escape is evoked by a different aversive stimulus electrical stimulation of the dPAG. Administration of the µ-receptor blocker CTOP into the dPAG did not change the escape threshold, but microinjection of the µ-receptor agonist DAMGO (0.3 and 0.5 nmol) or the 5-HT1A receptor agonist 8-OHDPAT (1.6 nmol) increased this index, indicating a panicolytic-like effect. Pretreatment with CTOP antagonised the anti-escape effect of 8-OHDPAT. Additionally, combined administration of subeffective doses of DAMGO and 8-OHDPAT increased the escape threshold, indicating drug synergism. Therefore, regardless of the aversive nature of the stimulus, µ-opioid and 5-HT1A receptors cooperatively act to regulate escape behaviour. A better comprehension of this mechanism might allow for new therapeutic strategies for panic disorder.

The median raphe nucleus in anxiety revisited.

Although the role of the median raphe nucleus (MRN) in the regulation of anxiety has received less attention than that of the dorsal raphe nucleus (DRN) there is substantial evidence supporting this function. Reported results with different animal models of anxiety in rats show that whereas inactivation of serotonergic neurons in the MRN causes anxiolysis, the stimulation of the same neurons is anxiogenic. In particular, studies using the elevated T-maze comparing serotonergic interventions in the MRN and in the DRN indicate that the former affect only the inhibitory avoidance task, which has been related to generalized anxiety. In contrast, similar operations in the DRN change both the inhibitory avoidance and the one-way escape task, the latter being representative of panic disorder. Simultaneous injections of 5-HT-acting drugs in the MRN and in the dorsal hippocampus (DH) suggest that the MRN-DH pathway mediates the regulatory function of the MRN in anxiety. Overall, the results discussed in this review point to a relevant role of the MRN in the regulation of anxiety, but not panic, through the 5-HT pathway that innervates the DH.

Cooperative regulation of anxiety and panic-related defensive behaviors in the rat periaqueductal grey matter by 5-HT1A and µ-receptors.

Previous results with the elevated T-maze (ETM) test indicate that the antipanic action of serotonin (5-HT) in the dorsal periaqueductal grey (dPAG) depends on the activation endogenous opioid peptides. The aim of the present work was to investigate the interaction between opioid- and serotonin-mediated neurotransmission in the modulation of defensive responses in rats submitted to the ETM. The obtained results showed that intra-dPAG administration of morphine significantly increased escape latency, a panicolytic-like effect that was blocked by pre-treatment with intra-dPAG injection of either naloxone or the 5-HT1A antagonist N-[2-[4-(2-methoxyphenyl)-1 piperazinyl] ethyl] -N- 2- pyridinyl-ciclohexanecarboxamide maleate (WAY-100635). In addition, previous administration of naloxone antagonized both the anti-escape and the anti-avoidance (anxiolytic-like) effect of the 5-HT1A agonist (±)-8-hydroxy-2-(di-n-propylamino)tetralin hydrobromide (8-OH-DPAT), but did not affect the anti-escape effect of the 5-HT2A agonist (±)-2,5-dimethoxy-4-iodoamphetamine hydrochloride (DOI). Moreover, the combination of sub-effective doses of locally administered 5-HT and morphine significantly impaired ETM escape performance. Finally, the µ-antagonist D-PHE-CYS-TYR-D-TRP-ORN-THR-PEN (CTOP) blocked the anti-avoidance as well as the anti-escape effect of 8-OHDPAT, and the association of sub-effective doses of the µ-opioid receptor agonist [D-Ala(2), N-Me-Phe(4), Gly(5)-ol]-enkephalin acetate salt (DAMGO) and of 8-OHDPAT had anti-escape and anti-avoidance effects in the ETM. These results suggest a synergic interaction between the 5-HT1A and the µ-opioid receptor at post-synaptic level on neurons of the dPAG that regulate proximal defense, theoretically related to panic attacks.

Anxiolytic-like effect of oxytocin in the simulated public speaking test.

Oxytocin (OT) is known to be involved in anxiety, as well as cardiovascular and hormonal regulation. The objective of this study was to assess the acute effect of intranasally administered OT on subjective states, as well as cardiovascular and endocrine parameters, in healthy volunteers (n = 14) performing a simulated public speaking test. OT or placebo was administered intranasally 50 min before the test. Assessments were made across time during the experimental session: (1) baseline (-30 min); (2) pre-test (-15 min); (3) anticipation of the speech (50 min); (4) during the speech (1:03 h), post-test time 1 (1:26 h), and post-test time 2 (1:46 h). Subjective states were evaluated by self-assessment scales. Cortisol serum and plasma adrenocorticotropic hormone (ACTH) were measured. Additionally, heart rate, blood pressure, skin conductance, and the number of spontaneous fluctuations in skin conductance were measured. Compared with placebo, OT reduced the Visual Analogue Mood Scale (VAMS) anxiety index during the pre-test phase only, while increasing sedation at the pre-test, anticipation, and speech phases. OT also lowered the skin conductance level at the pre-test, anticipation, speech, and post-test 2 phases. Other parameters evaluated were not significantly affected by OT. The present results show that OT reduces anticipatory anxiety, but does not affect public speaking fear, suggesting that this hormone has anxiolytic properties.

The panicolytic-like effect of fluoxetine in the elevated T-maze is mediated by serotonin-induced activation of endogenous opioids in the dorsal periaqueductal grey.

Serotonin (5-HT), opioids and the dorsal periaqueductal grey (DPAG) have been implicated in the pathophysiology of panic disorder. In order to study 5-HT-opioid interaction, the opioid antagonist naloxone was injected either systemically (1 mg/kg, i.p.) or intra-DPAG (0.2 µg/0.5 µL) to assess its interference with the effect of chronic fluoxetine (10 mg/kg, i.p., daily for 21 days) or of intra-DPAG 5-HT (8 µg/0.5 µL). Drug effects were measured in the one-escape task of the rat elevated T-maze, an animal model of panic. Pretreatment with systemic naloxone antagonized the lengthening of escape latency caused by chronic fluoxetine, considered a panicolytic-like effect that parallels the drug's therapeutic response in the clinics. Pretreatment with naloxone injected intra-DPAG antagonized both the panicolytic effect of chronic fluoxetine as well as that of 5-HT injected intra-DPAG. Neither the performance of the inhibitory avoidance task in the elevated T-maze, a model of generalized anxiety nor locomotion measured in a circular arena was affected by the above drug treatments. These results indicate that the panicolytic effect of fluoxetine is mediated by endogenous opioids that are activated by 5-HT in the DPAG. They also allow reconciliation between the serotonergic and opioidergic hypotheses of panic disorder pathophysiology.

Effects of diazepam on BOLD activation during the processing of aversive faces.

This study aimed to measure, using fMRI, the effect of diazepam on the haemodynamic response to emotional faces. Twelve healthy male volunteers (mean age = 24.83 ± 3.16 years), were evaluated in a randomized, balanced-order, double-blind, placebo-controlled crossover design. Diazepam (10 mg) or placebo was given 1 h before the neuroimaging acquisition. In a blocked design covert face emotional task, subjects were presented with neutral (A) and aversive (B) (angry or fearful) faces. Participants were also submitted to an explicit emotional face recognition task, and subjective anxiety was evaluated throughout the procedures. Diazepam attenuated the activation of right amygdala and right orbitofrontal cortex and enhanced the activation of right anterior cingulate cortex (ACC) to fearful faces. In contrast, diazepam enhanced the activation of posterior left insula and attenuated the activation of bilateral ACC to angry faces. In the behavioural task, diazepam impaired the recognition of fear in female faces. Under the action of diazepam, volunteers were less anxious at the end of the experimental session. These results suggest that benzodiazepines can differentially modulate brain activation to aversive stimuli, depending on the stimulus features and indicate a role of amygdala and insula in the anxiolytic action of benzodiazepines.

Reciprocal modulation of cognitive and emotional aspects in pianistic performances.

BACKGROUND: High level piano performance requires complex integration of perceptual, motor, cognitive and emotive skills. Observations in psychology and neuroscience studies have suggested reciprocal inhibitory modulation of the cognition by emotion and emotion by cognition. However, it is still unclear how cognitive states may influence the pianistic performance. The aim of the present study is to verify the influence of cognitive and affective attention in the piano performances. METHODS AND FINDINGS: Nine pianists were instructed to play the same piece of music, firstly focusing only on cognitive aspects of musical structure (cognitive performances), and secondly, paying attention solely on affective aspects (affective performances). Audio files from pianistic performances were examined using a computational model that retrieves nine specific musical features (descriptors)--loudness, articulation, brightness, harmonic complexity, event detection, key clarity, mode detection, pulse clarity and repetition. In addition, the number of volunteers' errors in the recording sessions was counted. Comments from pianists about their thoughts during performances were also evaluated. The analyses of audio files throughout musical descriptors indicated that the affective performances have more: agogics, legatos, pianos phrasing, and less perception of event density when compared to the cognitive ones. Error analysis demonstrated that volunteers misplayed more left hand notes in the cognitive performances than in the affective ones. Volunteers also played more wrong notes in affective than in cognitive performances. These results correspond to the volunteers' comments that in the affective performances, the cognitive aspects of piano execution are inhibited, whereas in the cognitive performances, the expressiveness is inhibited. CONCLUSIONS: Therefore, the present results indicate that attention to the emotional aspects of performance enhances expressiveness, but constrains cognitive and motor skills in the piano execution. In contrast, attention to the cognitive aspects may constrain the expressivity and automatism of piano performances.