Appropriate fluid intake is crucial to our vital functions. The negative effects of dehydration on not only physical1 but also cognitive performance have been reported.2,3 Using structural magnetic resonance imaging (MRI), Kempton et al.4 have shown that dehydration through physical exercise or restricted fluid intake causes reversible brain changes that consist of reduced brain volume and associated increases in ventricular volume. However, these effects of dehydration on pain thresholds and cortical activation in response to the pain experience remain unclear.
Pain is a conscious experience easily influenced by emotion, mood, and cognitive setting such as expectation, placebo, hypervigilance, attention, and distraction.5 Therefore, dehydration may affect the pain experience in humans. In previous studies, the effect of dehydration on human pain function, particularly on measures of brain function in humans, was not examined. Using functional MRI (fMRI), which allows indirect measurement of neural activity during event stimulus application, we specifically investigated the effects of dehydration on pain, compared with rehydration with an oral rehydration solution (ORS).
The aim of this study was to investigate the effects of dehydration on pain thresholds and cortical activation in response to evoked pain stimulus using fMRI and an executive function task (calculation test), compared with rehydration with ORS. At the central nervous system level, using fMRI, hemodynamic activity elicited by evoked pain stimulus in dehydrated subjects was compared with activity in rehydrated subjects given ORS. Blood-oxygen-level-dependent fMRI measures hemodynamic responses from changes in blood oxygenation, which are linked to changes in neural activity. To reproduce dehydration in healthy adult men in clear contrast to rehydration with ORS, the subjects performed 40 minutes of aerobic exercise in addition to 12-hour fasting without any oral rehydration. On a different day, subjects were rehydrated with ORS, although they performed an identical fasting and exercise routine. We hypothesized that pain experience, including pain thresholds and cortical activation, differs between dehydration and rehydration with ORS.
Five healthy, right-handed, neurologically normal men (mean age, 33.0 years; age range, 30–37 years), who provided their written informed consent, participated in this randomized, crossover, and repeated-measures design study (Fig. 1). This study was approved by the IRB of Gunma University Graduate School of Medicine, Maebashi, Japan and was conducted in accordance with institutional ethical provisions and the Declaration of Helsinki. The subjects were financially compensated for taking part in this study.
Conditions and Procedures
Participants were subjected to 2 separate hydration conditions on different days: dehydration day and rehydration day (Fig. 1). Each test day was held at intervals of >2 days to minimize the interaction between the conditions. Subjects underwent fMRI after exercise on either the dehydration or rehydration day. The subjects were instructed not to perform strenuous activity or consume alcohol. The condition on the first day was randomly assigned to each subject by tossing a coin.
On the dehydration day, subjects were instructed to refrain from eating and drinking for 12 hours before fMRI. On the rehydration day, subjects were instructed to fast for 12 hours similar to the dehydration condition, although they were instructed to consume up to 3000 mL (up to 2000 mL from the night before the test day until the time they started exercise and up to 1000 mL during exercise) ORS containing balanced amounts of glucose and electrolytes (OS-1®, [http://http://www.otsukakj.jp/en/profile/products/mf/os1.html], Otsuka Pharmaceutical Factory, Inc., Tokushima, Japan). Immediately before fMRI, they reported the total amount consumed since the night before.
Physiological Measures, Subjective State Measures, and Cognitive Test
Subjects gave their physiological and subjective state measures twice: before and after exercise (Fig. 1). On arrival at the laboratory, subjects first underwent a medical assessment, comprising initial assessment of clinical history, including details of any current or previous medical or surgical morbidity, medications taken, recent substance use, and allergies. All subjects were determined to be fit and healthy.
This was followed by general physical and physiological examinations including those of basic measures (nude body weight [accurate to 50 g, TANITA, Tokyo, Japan], arterial blood pressure, heart rate, and tympanic temperature). These examinations were also performed after exercise as a postexercise medical check. After exercise, the subjects provided a urine sample for determination of urine osmolality (mOsm/kg) (OM-6050, ARKRAY Inc., Kyoto Japan).
After physical and physiological examinations, to measure the subjective mood index of dehydration and rehydration with ORS, subjects rated the following 3 items on a 5-point scale questionnaire: thirst, hunger, and anxiety levels, rating them from 1 (not at all) to 5 (very). Similarly, the subjects rated their mood using the visual analog mood scale, which is a 100-mm horizontal line and simple, schematic faces representing mood states from 0 (worst imaginable health state) to 100 (best imaginable health state), first described by Aitken,6 which has been shown to be a reliable and valid measure of mood in psychiatric populations.7
After exercise, subjects were given a 10-minute supine recovery period. Then, for a cool-down period of 50 minutes before fMRI, subjects underwent physiological and subjective measurements (30 minutes) and the Uchida-Kraepelin Performance test (20 minutes), which is a simple arithmetic test that measures task performance speed and task performance accuracy. In this test, subjects were instructed to add 2 single digits and answer only using single digits as fast as possible, and the results were evaluated from the number of calculations and error rate. We used the Uchida-Kraepelin Performance test (Nisseiken, Inc., Tokyo, Japan) to examine the effects of dehydration on executive cognitive functions.8
Both conditions included identical 40-minute exercise in a training room in the laboratory. During the exercise, subjects wore light sports clothing (t-shirt, shorts, socks, and athletic shoes). The exercise protocol was identical for both conditions and was divided into two 20-minute sessions for a total of 40 minutes involving elliptical walking exercise on a total body elliptical walking machine, Reebok Body Trec Elliptical Trainer® (Reebok International Ltd., Canton, MA), in the “steady-climb” mode starting at level 3 (load level varied from 3 to 10), which was adjusted depending on individual physical pace. We gave subjects a 5-minute rest between the two 20-minute sessions. With elliptical trainers, the subjects stand on raised pedals and hold bars parallel to the machine, moving their arms forward and back while their legs move in an elliptical motion. The amount of exercise was determined using the estimated calories burned while exercising on the basis of elliptical walking machine algorithms using subject age, body weight, pace, and time. For the subjects’ safety, 1 doctor or nurse was assigned to each subject during exercise, monitoring his health state constantly; the training room was also equipped with an automatic external defibrillator and a resuscitation kit.
Pain and Control Stimuli During fMRI Scanning
During fMRI scanning, 2 different pressure tests were used: the cold pressor test (CPT) for generating experimental pain and a test using a nonpainful pressure (as a control stimulus), perceived only as a light touch. CPT has been established as an easy, safe, and reliable method of pain induction using cold water that evokes cold pain.9,10 Previously, we used a CPT by applying frozen ice packs (size, 9 × 9 cm, weight 37 g) on each subject’s medial forearm to eliminate the need for large metal water baths in the MRI room and to avoid drenching the MRI gantry.11 We applied the same packs as a control stimulus on each subject’s medial forearm, but the packs were not frozen and were swathed in cotton tissue to approximate a light touch. We called this control stimulus “control” in this study. Each subject underwent a total of 20 blocks (CPT, 10 blocks; control, 10 blocks). The order of presentation of all stimulation blocks was randomized over the scanning time, preventing subjects from anticipating the event type, counterbalanced to be the same number between the CPT and control blocks.
Pain Threshold Measurements
In a CPT block, we measured a subject’s pain threshold (defined as the time when the subject started to feel pain induced by an ice pack). In accordance with the experimental program, the experimenter applied an ice pack (frozen, CPT block; not frozen, control block) to a 15 × 5 cm2 stimulation area marked on the subject’s medial forearm, alternating between right and left arms in each block. In a CPT block, the subject pressed a button with his nonstimulated hand at the onset of pain; the time was recorded on the computer log as the pain threshold, which is the short-term pain threshold in CPT.12 Twenty seconds after the start of CPT, the experimenter removed the ice pack. For each subject’s safety, we limited the CPT time to 20 seconds regardless of whether the subject reported a pain threshold. fMRI scanning was started before the start of the CPT (10 scans used as the baseline) and lasted 21 seconds after the start of the CPT (7 scans). Between blocks, we provided a 2-minute interval to minimize the effect of the preceding stimulus, during which the subject’s stimulated forearm was swaddled with a towel to minimize sensitization to the next stimulus.
MRI was performed using a 3.0-tesla Siemens MAGNETOM Trio scanner (MAGNETOM Trio A, Tim System; Siemens, Malvern, PA) at the Brain Activity Imaging Center (Kyoto, Japan). The acquired functional images consisted of 50 consecutive slices parallel to the anterior and posterior commissure planes, covering the whole brain. A T2*-weighted gradient-echo echo-planar imaging sequence was used with the following parameters: repetition time = 3000; echo time = 30 milliseconds; flip angle = 90°; matrix size = 64 × 64; voxel size = 3 × 3 × 3 mm3. After acquisition of functional images, T1-weighted high-resolution anatomical images were obtained using a magnetization-prepared rapid acquisition gradient-echo sequence (repetition time = 2250 milliseconds; echo time = 3.06 milliseconds; flip angle = 9°; field of view = 256 × 256 mm2; voxel size = 1 × 1 × 1 mm3).
Physiological and Subjective Measurements
All subjects’ physiological data including pain thresholds, subjective data (i.e., thirst, hunger, anxiety levels, and mood) and the results of the Uchida-Kraepelin performance test were compared between dehydration and rehydration using a 1-sample t test. A significance level of
P < 0.05 was used.
Image and statistical analyses were performed using the statistical parametric mapping package SPM8 (http://www.fil.ion.ucl.ac.uk/spm) implemented in MATLAB R2007a (Mathworks Inc., Natick, MA). Functional images in each run were realigned using the first scan as a reference to correct for head movements. Data from all 5 subjects required a small-motion correction (<2 mm). The obtained T1-weighted anatomical images were preprocessed by intensity inhomogeneity correction. The T1-weighted anatomical images were then coregistered to the first scan of the functional images. After this, the coregistered T1 anatomical images were normalized to a standard T1 template image, as defined by the Montreal Neurological Institute involving linear and nonlinear 3-dimensional transformations.13,14 The parameters from this normalization process were then applied to each of the functional images. Finally, these spatial normalized functional images were resampled to a voxel size of 2 × 2 × 2 mm3 and smoothed with an isotropic Gaussian kernel of 8 mm full width at half maximum to compensate for anatomic variability among subjects.
We used random effect analyses15 to assess the statistical effects of stimulus type (CPT or control) under each condition (dehydration or rehydration with ORS). Data were first analyzed for each subject (a single-subject analysis) and then for the group. Random effect analysis considers the between-subject variability of the responses to allow for population inference. In SPM, random effect analysis can be implemented using a 2 -level summary statistics approach as follows. At the first level, individual brain responses under each condition are estimated using subject-specific statistical models, which include regressors of experimental effects (i.e., task-related neural activities) and additional nuisance variables (e.g., signal drift). These individual brain responses are then entered into a second-level analysis for a statistical test across the entire sample, thus allowing for population inference.
In our study, we specified the following model in the first-level analysis: the task-related (CPT or control) neural activity under each condition (dehydration or rehydration with ORS) was modeled with a boxcar function, which was convoluted with a canonical hemodynamic response function. Note that a canonical (or standard) hemodynamic response function is typical of an evoked blood-oxygen-level-dependent response to brief neural stimulation. The canonical hemodynamic response function used in SPM software shows a rising peak at approximately 6 seconds followed by an undershoot.15 We used a high-pass filter to remove low-frequency signal drift. Serial autocorrelation, assuming a first-order autoregressive model, was also corrected.
In the second-level analysis, we used repeated-measures analysis of variance (ANOVA) using the full factorial design in SPM second-level statistics that tests each contrast against a pooled error term (with 16 degrees of freedom [df]), which can improve the normality of the statistics even with the use of few samples.16 Note that the number of available df for an error term is 16 since there are 4 regressors for task conditions in our model with 4 df and hence leaving 16 df from 20 images (4 task conditions × 5 subjects). The subject-specific contrast images for dehydration and rehydration conditions were subjected to a 2 (stimulus type: CPT or control) × 2 (condition: dehydration or rehydration with ORS) ANOVA. Finally, we tested the following 4 contrasts of brain activities: (1) [CPT−control (dehydration)], CPT minus control stimulus during dehydration; (2) [CPT−control (rehydration)], CPT minus control stimulus during rehydration with ORS; (3) [control−CPT (dehydration)], control stimulus minus CPT during dehydration; 4) [control−CPT (rehydration)], control stimulus minus CPT during rehydration with ORS. Since we had no a priori predictions about the location of effects, we used a corrected threshold for multiple comparisons, which is commonly used in neuroimaging data analysis to search for significantly activated voxels across the whole brain on the basis of a random field theory.17
Significantly activated voxels were identified when they reached the extent threshold of P < 0.05 corrected for multiple comparisons, with a height threshold of P < 0.001 (uncorrected) across the whole brain. Brain activation is inherently a statistical image where each voxel contains 1 statistical value (i.e., t value) thresholded by some height and spatial extent thresholds that are specified by the experimenter. After thresholding, we can define significant topological features such as peak and cluster size. In Table 2, we present the cluster-level corrected P: the significant P-value for a cluster (quantified by the number of voxels) at a threshold of P < 0.05 after the correction of multiple comparisons, and t-values for a significant activation peak when they are above the threshold of P < 0.001.
Because of the small number of subjects, we also confirmed the consistency of effects across all subjects by conjunction analysis with a global null hypothesis18,19 instead of the second-level random effect analysis to validate our approach. Results are shown in Figure 4.
Physiological Measures and Pain Threshold Measurement Results
On the rehydration day, subjects consumed an average of 2040 mL (range 1800–2500 ml) ORS. Table 1 shows a summary of physiological and subjective data for both conditions (dehydration and rehydration) including pain thresholds and Uchida-Kraepelin performance test results.
A 1-sample t test of physiological data revealed that physiological measurements showed statistically significant differences between the dehydrated and rehydrated subjects (Table 1). The pain threshold during dehydration was 13.4 ± 3.6 seconds (mean ± standard deviation: SD), and during rehydration was 16.2 ± 3.4 seconds (mean ± SD) (Fig. 2). During dehydration, a significant decrease in pain threshold was observed compared with that during rehydration with ORS.
Subjective State Measures and Cognitive Test Results
Subjective state measures and cognitive test results (the lower section of Table 1) revealed that subjects expressed a significantly stronger thirst while dehydrated than while rehydrated, although hunger and anxiety levels and mood did not significantly differ between conditions. An executive function task (Uchida-Kraepelin Performance test) performed after exercise revealed that dehydration is associated with a significantly smaller amount of work than rehydration, although the error rate did not significantly differ between conditions.
From the 2 × 2 ANOVA results including brain images from CPT or control (2 × stimulus type) from dehydration and rehydration (2 × condition), [CPT−control] contrasts during both dehydration and rehydration conditions revealed a comprehensive representation of the pain network during CPT. Regions such as the anterior cingulate cortex (ACC), bilateral insula, thalamus, and cerebellum were activated. The overall activation during dehydration was greater than that during rehydration with ORS in terms of peak and cluster (Fig. 3, A and B and Table 2).
However, in the [control−CPT] contrasts under both conditions, there were significant clusters located in the ventromedial prefrontal cortex (VMPFC), posterior cingulate cortex, and amygdala. The overall activation during rehydration with ORS was greater than during dehydration in terms of peak and cluster (Fig. 3, C and D and Table 2). In Table 2, the activated brain regions in each contrast are listed with their numbers of voxels, cluster-level corrected P-values, coordinates, and t-values at the voxel level.
This study shows that dehydrated subjects exhibit increased pain-related brain activity in the ACC and insula during pain processing with decreased pain thresholds, compared with rehydrated subjects.
This study had the advantages of measuring supraspinal responses to painful stimuli and comparing the responses between dehydration and rehydration with ORS, as well as differences in pain threshold and thirst level. There was no previous study that used neuroimaging to investigate the effects of dehydration on pain perception. In 1 study, however, the effects of dehydration on brain function were examined by fMRI in healthy adolescents, demonstrating attenuated executive functions such as planning and visuospatial processing during dehydration.20 In our executive function task (Uchida-Kraepelin performance test) after exercise, dehydration led to a significantly smaller amount of work than rehydration with ORS, although the error rate did not significantly differ between conditions (Table 1). In a previous study,20 poor executive function task performance validated the notion that dehydration can lead to negative cognitive changes. Analogous findings of cognitive function impairment during dehydration have been reported.21,22
There are possible neural mechanisms by which dehydration affects pain experience negatively. This study shows clear differences in physiological data between dehydration and rehydration with ORS (Table 1). Subjects reported significantly strong thirst while dehydrated. Moreover, they did not report thirst while rehydrated, and there were no significant differences in other subjective measurements such as hunger, anxiety levels, and general mood. The only significant difference in subjective state measures between dehydration and rehydration with ORS was in thirst level (Table 1).
We, therefore, suggest that the key factor is the subjective feeling of “thirst,” which leads to differences in pain-related brain activity between dehydration and rehydration with ORS (Fig. 3, A and B). Thirst is one of the most powerful sensations driving the need for moisture for survival.23,24 Intriguingly, Egan et al.25 using fMRI showed high activities in the ACC, insula, and cerebellum in thirsty human subjects. They showed that the ACC might be the region for consciousness of thirst, because when subjects were rehydrated with water, an immediate decrease in activity occurred in the ACC with the attenuation of thirst.25
When humans are dehydrated, it is not surprising that such a strong motivational sensation called thirst amplifies the pain-related cerebral activity, because pain is an unpleasant sensation. However, it is always subjective and associated with emotions.26 A number of studies have suggested that negative conditions and emotion can lead to pain or pain exacerbation.27,28 It seems likely that the ACC and insula are actually part of a multimodal network related to the detection of multiple salient sensory inputs.29 Our finding of increased pain-related cerebral activity during dehydration might be exacerbated by thirst, which threatens our lives. There is evidence of such a causal relationship in which negative conditions and emotions can lead to pain or pain exacerbation.30
During rehydration with ORS, however, together with prolonged pain thresholds, our brain data on [control−CPT] contrasts demonstrate increased reward circuitry activity in regions such as the VMPFC and amygdala (Fig. 3, C and D), which are thought to be major dopaminergic targets and have been implicated in the reward network.31 These denote significantly greater brain activation in response to light-touch stimulus (control stimulus) than in response to a pain stimulus (CPT). Generally, the brain’s reward network activation is associated with pleasure, involving the VMPFC, amygdala, and ventral tegmental area, which respond to not only physical rewards such as food, drinks, drugs, and sexual activity but also social rewards such as having a good reputation32 or fair treatment.33
As mentioned above, we suggest that dehydration generates strong thirst, leading to an unstable state and sensitivity of subjects to painful stimuli, as represented by markedly high pain-related brain activity. Thus, the process of satisfying strong thirst must be pleasurable and rewarding (e.g., a cold beverage tastes better with a dry throat), as demonstrated by the increased reward circuitry brain activity in our rehydrated subjects. At the very least, thirst during dehydration might be a strong physical survival need; therefore, once subjects were rehydrated with ORS, the difference between the conditions with or without thirst may have affected the pain experience.
The mechanisms underlying the modulation of pain by emotion mainly involve the top-down pain modulatory system comprising the prefrontal cortex, amygdala, hypothalamus, and brainstem structures such as the periaqueductal grey and the descending projections to the spinal dorsal horn,34,35 although we found no significant activation below the brainstem level or in the prefrontal region. However, Takahashi et al.36 first demonstrated a correlative interplay between pain-related brain activation and reward network activation in their fMRI study. Hence, a feeling of thirst and its alleviation might have a considerable effect on pain experience by such a correlative interplay between dehydration (painful experience) and rehydration with ORS (reward), as mentioned above.
First, the small number of subjects is a limitation of this study because statistical power depends on total sample size, and thus, it is desirable to have as many subjects as possible.37 Therefore, more subjects are necessary for future studies. Second, the control group (rehydration with ORS in this study) did not include a real control group such as a group rehydrated with water without fasting or exercise, which would be a better control group. Third, a simple 5-point scale measurement method for anxiety level and mood using the visual analog mood scale might contribute to the absence of significant differences in anxiety level and mood between the conditions used in this study. Valid and specific tools for measuring anxiety (e.g., State-Trait Anxiety Inventory) may be ideal for future assessments. Last, the increased activity of the reward network shown during rehydration with ORS conditions is an intriguing observation in this study. However, it is difficult to speculate about the cause of this increased activity of the reward network and attenuated pain during rehydration with ORS. Difficulties in interpreting such interactions are also limitations of this study.
This study shows that dehydration increases brain activity related to painful stimuli with decreased pain threshold and enhanced thirst in humans, whereas rehydration with ORS alleviates thirst and decreases this brain activity.
Name: Yuichi Ogino, MD, PhD.
Contribution: This author unified design and conduct of the study, data collection, data analysis, and manuscript preparation. This author is the first and corresponding author.
Attestation: Yuichi Ogino is the principal author and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Takahiro Kakeda, RN, PHN, PhD.
Contribution: This author helped in the conduct of the study, data collection, and data analysis.
Attestation: Takahiro Kakeda approved the final manuscript.
Name: Koji Nakamura, MD.
Contribution: This author helped in design and conduct of the study, data collection, and data analysis.
Attestation: Koji Nakamura approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Shigeru Saito, MD, PhD.
Contribution: This author helped in study design and preparation of the manuscript.
Attestation: Shigeru Saito approved the final manuscript.
This manuscript was handled by: Spencer S. Liu, MD.
We are very grateful to Koji Inui, MD, PhD, associate professor of the Department of Integrative Physiology, National Institute for Physiological Sciences (Okazaki, Japan) for fruitful discussion and comments. We are also grateful to Takanori Kochiyama, PhD, and Yukiko Nota, PhD, of the Brain Activity Imaging Center, Advanced Telecommunications Research Institute International (Kyoto, Japan) for technical advice and supporting this project.
1. Duning T, Kloska S, Steinsträter O, Kugel H, Heindel W, Knecht S. Dehydration confounds the assessment of brain atrophy. Neurology. 2005;64:548–50
2. Wilson MM, Morley JE. Impaired cognitive function and mental performance in mild dehydration. Eur J Clin Nutr. 2003;57(Suppl 2):S24–9
3. Cian C, Barraud PA, Melin B, Raphel C. Effects of fluid ingestion on cognitive function after heat stress or exercise-induced dehydration. Int J Psychophysiol. 2001;42:243–51
4. Kempton MJ, Ettinger U, Schmechtig A, Winter EM, Smith L, McMorris T, Wilkinson ID, Williams SC, Smith MS. Effects of acute dehydration on brain morphology in healthy humans. Hum Brain Mapp. 2009;30:291–8
5. Tracey I, Mantyh PW. The cerebral signature for pain perception and its modulation. Neuron. 2007;55:377–91
6. Aitken RC. Measurement of feelings using visual analogue scales. Proc R Soc Med. 1969;62:989–93
7. Bech P, Kastrup M, Rafaelson OJ. Mini-compendium of rating scales for states of anxiety, depression, mania, schizophrenia with corresponding DSM-III syndromes Acta Psychiatrica Scandinavica. 1986;73(Suppl. 326):7–37
8. Hamazaki-Fujita N, Hamazaki K, Tohno H, Itomura M, Terashima Y, Hamazaki T, Nakamura N, Yomoda S. Polyunsaturated fatty acids and blood circulation in the forebrain during a mental arithmetic task. Brain Res. 2011;1397:38–45
9. Posner J, Telekes A, Crowley D, Phillipson R, Peck AW. Effects of an opiate on cold-induced pain and the CNS in healthy volunteers. Pain. 1985;23:73–82
10. Chen AC, Dworkin SF, Haug J, Gehrig J. Human pain responsivity in a tonic pain model: psychological determinants. Pain. 1989;37:143–60
11. Kakeda T, Ogino Y, Moriya F, Saito S. Sweet taste-induced analgesia: an fMRI study. Neuroreport. 2010;21:427–31
12. Rainville P, Doucet JC, Fortin MC, Duncan GH. Rapid deterioration of pain sensory-discriminative information in short-term memory. Pain. 2004;110:605–15
13. Ashburner J, Friston KJ. Nonlinear spatial normalization using basis functions. Hum Brain Mapp. 1999;7:254–66
14. Friston KJ, Holmes AP, Poline JB, Grasby PJ, Williams SC, Frackowiak RS, Turner R. Analysis of fMRI time-series revisited. Neuroimage. 1995;2:45–53
15. Friston KJ, Ashburner JT, Kiebel SJ, Nichols TE, Penny WD Statistical Parametric Mapping: The Analysis of Functional Brain Images. 2007 London Elsevier Academic Press
16. Luo WL, Nichols TE. Diagnosis and exploration of massively univariate neuroimaging models. Neuroimage. 2003;19:1014–32
17. Worsley KJ, Marrett S, Neelin P, Vandal AC, Friston KJ, Evans AC. A unified statistical approach for determining significant signals in images of cerebral activation. Hum Brain Mapp. 1996;4:58–73
18. Friston KJ, Holmes AP, Worsley KJ. How many subjects constitute a study? Neuroimage. 1999;10:1–5
19. Friston KJ, Penny WD, Glaser DE. Conjunction revisited. Neuroimage. 2005;25:661–7
20. Kempton MJ, Ettinger U, Foster R, Williams SC, Calvert GA, Hampshire A, Zelaya FO, O’Gorman RL, McMorris T, Owen AM, Smith MS. Dehydration affects brain structure and function in healthy adolescents. Hum Brain Mapp. 2011;32:71–9
21. McMorris T, Swain J, Smith M, Corbett J, Delves S, Sale C, Harris RC, Potter J. Heat stress, plasma concentrations of adrenaline, noradrenaline, 5-hydroxytryptamine and cortisol, mood state and cognitive performance. Int J Psychophysiol. 2006;61:204–15
22. Baker LB, Conroy DE, Kenney WL. Dehydration impairs vigilance-related attention in male basketball players. Med Sci Sports Exerc. 2007;39:976–83
23. Parsons LM, Denton D, Egan G, McKinley M, Shade R, Lancaster J, Fox PT. Neuroimaging evidence implicating cerebellum in support of sensory/cognitive processes associated with thirst. Proc Natl Acad Sci U S A. 2000;97:2332–6
24. Denton D, Shade R, Zamarippa F, Egan G, Blair-West J, McKinley M, Lancaster J, Fox P. Neuroimaging of genesis and satiation of thirst and an interoceptor-driven theory of origins of primary consciousness. Proc Natl Acad Sci U S A. 1999;96:5304–9
25. Egan G, Silk T, Zamarripa F, Williams J, Federico P, Cunnington R, Carabott L, Blair-West J, Shade R, McKinley M, Farrell M, Lancaster J, Jackson G, Fox P, Denton D. Neural correlates of the emergence of consciousness of thirst. Proc Natl Acad Sci U S A. 2003;100:15241–6
26. Fields HL. Pain: an unpleasant topic. Pain. 1999(Suppl 6):S61–9
27. Ploner M, Lee MC, Wiech K, Bingel U, Tracey I. Flexible cerebral connectivity patterns subserve contextual modulations of pain. Cereb Cortex. 2011;21:719–26
28. Berna C, Leknes S, Holmes EA, Edwards RR, Goodwin GM, Tracey I. Induction of depressed mood disrupts emotion regulation neurocircuitry and enhances pain unpleasantness. Biol Psychiatry. 2010;67:1083–90
29. Iannetti GD, Mouraux A. From the neuromatrix to the pain matrix (and back). Exp Brain Res. 2010;205:1–12
30. Wiech K, Tracey I. The influence of negative emotions on pain: behavioral effects and neural mechanisms. Neuroimage. 2009;47:987–94
31. Berridge KC, Kringelbach ML. Affective neuroscience of pleasure: reward in humans and animals. Psychopharmacology (Berl). 2008;199:457–80
32. Izuma K, Saito DN, Sadato N. Processing of social and monetary rewards in the human striatum. Neuron. 2008;58:284–94
33. Tabibnia G, Satpute AB, Lieberman MD. The sunny side of fairness: preference for fairness activates reward circuitry (and disregarding unfairness activates self-control circuitry). Psychol Sci. 2008;19:339–47
34. Kong J, Gollub RL, Rosman IS, Webb JM, Vangel MG, Kirsch I, Kaptchuk TJ. Brain activity associated with expectancy-enhanced placebo analgesia as measured by functional magnetic resonance imaging. J Neurosci. 2006;26:381–8
35. Bantick SJ, Wise RG, Ploghaus A, Clare S, Smith SM, Tracey I. Imaging how attention modulates pain in humans using functional MRI. Brain. 2002;125:310–9
36. Takahashi H, Kato M, Matsuura M, Mobbs D, Suhara T, Okubo Y. When your gain is my pain and your pain is my gain: neural correlates of envy and schadenfreude. Science. 2009;323:937–9
37. Desmond JE, Glover GH. Estimating sample size in functional MRI (fMRI) neuroimaging studies: statistical power analyses J Neurosci Methods. 2002;118:115–28