R

Cc

Aa

Tb

o

a

ARRA

KCCAANAANNN

C

MSf

0h

Neuroscience and Biobehavioral Reviews 37 (2013) 2445–2453

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews

jou rn al h om epage: www.elsev ier .com/ locate /neubiorev

eview

hocolate and the brain: Neurobiological impact of cocoa flavanols onognition and behavior

lexander N. Sokolova,∗, Marina A. Pavlovab, Sibylle Klosterhalfena, Paul Encka

Department of Internal Medicine VI: Psychosomatic Medicine and Psychotherapy, Research Division, University of Tübingen Medical School, D 72076übingen, GermanyDevelopmental Cognitive and Social Neuroscience Unit, Department of Pediatric Neurology and Developmental Medicine, Children’s Hospital, Universityf Tübingen Medical School, D 72076 Tübingen, Germany

r t i c l e i n f o

rticle history:eceived 18 February 2013eceived in revised form 17 June 2013ccepted 18 June 2013

eywords:ocoa flavanolshocolatentioxidantnti-inflammatory

a b s t r a c t

Cocoa products and chocolate have recently been recognized as a rich source of flavonoids, mainly fla-vanols, potent antioxidant and anti-inflammatory agents with established benefits for cardiovascularhealth but largely unproven effects on neurocognition and behavior. In this review, we focus on neuro-modulatory and neuroprotective actions of cocoa flavanols in humans. The absorbed flavonoids penetrateand accumulate in the brain regions involved in learning and memory, especially the hippocampus. Theneurobiological actions of flavanols are believed to occur in two major ways: (i) via direct interactionswith cellular cascades yielding expression of neuroprotective and neuromodulatory proteins that pro-mote neurogenesis, neuronal function and brain connectivity, and (ii) via blood-flow improvement andangiogenesis in the brain and sensory systems. Protective effects of long-term flavanol consumption

eurogenesisngiogenesisge- and disease-related declineeurocognitioneuromodulationeuroprotection

on neurocognition and behavior, including age- and disease-related cognitive decline, were shown inanimal models of normal aging, dementia, and stroke. A few human observational and intervention stud-ies appear to corroborate these findings. Evidence on more immediate action of cocoa flavanols remainslimited and inconclusive, but warrants further research. As an outline for future research on cocoa flavanolimpact on human cognition, mood, and behavior, we underscore combination of functional neuroimagingwith cognitive and behavioral measures of performance.

© 2013 Published by Elsevier Ltd.

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24452. Cocoa flavanols in the brain signaling cascades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24463. Neuroprotective action of cocoa flavanols in aging and neurological disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2447

3.1. Animal studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24473.2. Human population studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24483.3. Clinical studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2448

4. Neuromodulation of cognition, mood, learning, and memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24494.1. Animal studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24494.2. Human studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2449

5. Concluding remarks and future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2450Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2452References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2452

∗ Corresponding author at: Department of Internal Medicine VI: Psychosomaticedicine and Psychotherapy, Research Division, University of Tübingen Medical

chool, Osianderstr. 5, D 72076 Tübingen, Germany. Tel.: +49 7071 29 89118;ax: +49 7071 29 4382.

E-mail address: alexander.sokolov@uni-tuebingen.de (A.N. Sokolov).

149-7634/$ – see front matter © 2013 Published by Elsevier Ltd.ttp://dx.doi.org/10.1016/j.neubiorev.2013.06.013

1. Introduction

Cocoa products and especially, chocolate has taken a specialplace in our daily life and culture. This food of the gods as tells its

Latin name Theobroma cacao given by the noted Swedish nosologistCarl Linnaeus in 1753, has been ennobled in many countries aroundthe globe as a curative drug, a culinary delight, and even a currencyfor commodity trading, retaining its appeal over the centuries. No

2 iobeh

oaie(hphoopbf“

ccaqaDflpt(2nyeMcN

efetKh2dtiiPtatfarnitb

2

t2Tovht

446 A.N. Sokolov et al. / Neuroscience and B

ther natural product but chocolate has ever been viewed as having positive effect on a wide variety of health conditions ranging fromntestinal and female complaints, fever, and cardiovascular dis-ases to promotion of strength before military and sexual conquestsWilson, 2010; Wolfe and Shazzie, 2005). Reports on chocolate’sealth benefits are dated back as far as Aztec and Maya medicalractice (e.g., Hurst et al., 2002) and ever since, anecdotal evidenceas been abundant on chocolate effects on health. Only by the endf the 20th century, however, claims on supposed health benefitsf chocolate have increasingly drawn a scientific interest in cocoaroducts and chocolate, which eventually resulted in an approvaly the European Food Safety Agency (EFSA, 2012) of a health claimor dark chocolate with high flavanol content as to its impact onmaintenance of normal endothelium-dependent vasodilation”.

Most studies so far have been conducted on the effects ofhocolate intake on the cardiovascular system, skin, cholesteroloncentrations, and the release of neurotransmitters anandamidend serotonin, and on the health-related properties of high-uality dark chocolate, containing the stimulants theobrominend caffeine (Lamuela-Raventós et al., 2005; Katz et al., 2011).ark chocolate also comprises high concentrations of flavanols (aavonoid subgroup, mainly epicatechin; Whiting, 2001) known asotent antioxidative agents. While some work has been done onhe influence of theobromine and caffeine on mood and cognitione.g., Smit and Rogers, 2000; Smit et al., 2004; Smit and Blackburn,005; Nehlig, 2010), the impact of cocoa flavanols on human cog-itive and affective function, executive control and behavior haset to be determined. In accord with accumulating evidence fornhancing effects of chocolate consumption on cognitive function,esserli (2012) reports, as an occasional note, a strong positive

orrelation between chocolate intake per capita and the number ofobel laureates in various countries.

In contrast to potential effects on cognition and behavior,vidence-based benefits of cocoa and chocolate consumptionor cardiovascular system are well established and includendothelium-dependent vasodilation recently found to contributeo normal blood flow (Engler et al., 2004; Hooper et al., 2012;ay et al., 2006; Grassi et al., 2005, 2008). Cardiovascular healthas been closely linked to cognitive performance (e.g., DeCarli,012). Animal studies have shown that the absorbed flavonoidsirectly interact with a number of cellular and molecular targets inhe brain, exerting pronounced antioxidative effects and improv-ng brain tissue and function in the regions mainly implicatedn learning, memory, and cognition (Andrés-Lacueva et al., 2005;assamonti et al., 2005; Vauzour et al., 2008). This suggests a poten-ial neuromodulatory and neuroprotective role for cocoa flavanolsnd their significance for cognitive and affective function, execu-ive control and behavior. However, only few human studies soar have specifically addressed neurobiological, cognitive, affectivend behavioral effects of flavanol-rich cocoa products. The presenteview focuses on analysis of the existing evidence on potentialeuromodulatory and neuroprotective actions of cocoa flavanols

n humans. Our analysis highlights further ways in investigation ofhe impact of flavanol-rich cocoa products on neurocognition andehavior.

. Cocoa flavanols in the brain signaling cascades

The flavanol monoisomers epicatechin and catechin arehe predominant flavonoid compounds in cocoa, with the-phenyl-3,4-dihydro-2H-chromen-3-ol as underlying skeleton.hese monomers represent the base molecules for concatenated

ligomers, the proanthocyanidins. Antioxidant properties of fla-anols are chemically mediated through oxidation of two aromaticydroxyl groups to a quinone (Bors and Michel, 2002). In addi-ion, flavanols foster antioxidant system through modulation of

avioral Reviews 37 (2013) 2445–2453

enzymatic activity (Stevenson and Hurst, 2007; Mann et al., 2009).Flavanols occur in high concentrations in beverages such as greentea and red wine, fruits and berries (e.g., apple skin, grapes, pears,blueberry), vegetables (tomatoes, soy, and olives), and, especially,cocoa (Manach et al., 2004; Neveu et al., 2010; Scalbert et al.,2011; Sies et al., 2012). The flavonoid contents in cocoa prod-ucts and chocolate differ substantially depending on the cocoavariety (in some beans, amounting up to 20%), geographic origin,cultivating, agricultural and postharvest practices, and manufac-turing (Wollgast and Anklam, 2000; Niemenak et al., 2006). In theDutch population, chocolate contributes to about 20% of the totalflavonoid intake in adults, with an even higher percentage in chil-dren (Arts et al., 1999). In American diet, chocolate represents thethird top source of antioxidants after fruits and vegetables (255,233, and 104 mg/day, respectively; Vinson et al., 2006). Also in theFrench adult population with the total daily dietary polyphenolintake of 1.2 g, including 99 mg catechins, cocoa products accountfor the third major source of epicatechin (17%) after green tea (28%)and apples (24%; Pérez-Jiménez et al., 2011). Yet human and animalstudies on neuroprotective properties of flavonoids, especially inpreventing cognitive decline, have mostly examined plant-derivedsubstances other than cocoa flavanols (e.g., Macready et al., 2009).Flavonoid-rich pine extracts such as Gingko biloba are reported todelay the onset of memory loss, dementia, and Alzheimer’s disease(Weinmann et al., 2010), but the evidence remains controversial(Birks and Grimley Evans, 2009).

Animal studies show that flavanols and their metabolites cancross the blood–brain barrier, inducing beneficial effects on thebrain tissue and function (angio- and neurogenesis, changes in neu-ron morphology) and stimulating widespread blood circulation inthe brain (Vauzour et al., 2008). The most common flavanol found incocoa, epicatechin (Whiting, 2001), is rapidly absorbed in humansand is detectable in blood plasma already 30 min after intake. Theepicatechin levels peak 2–3 h after intake, exhibiting a strong posi-tive correlation with the dose of ingested chocolate (Richelle et al.,1999), and return to baseline by 6–8 h after consumption. The pos-sibility of flavanols and metabolites to penetrate and accumulatein the brain regions mainly related to learning and memory, sug-gests they may exert a direct positive impact on the brain, includingcognition and neuroprotection (Nehlig, 2013).

Neurobiological impact of flavanols on the brain, learning, mem-ory, and cognition are believed to occur in two major ways (Fig. 1).First, flavonoids can specifically interact within a number of cellu-lar signaling pathways, primarily with mitogen-activated protein(MAPK), extracellular-signal-regulated (ERK) and phosphoinosi-tide 3-kinase (PI3-kinase/Akt) signaling cascades. These cascadestrigger gene expression and protein synthesis for maintaininglong-term potentiation (LTP) and establishing long-term memo-ries (Kelleher et al., 2004). Flavonoids modulate the transcriptionfactors engaged in signal transduction through protein-kinase inhi-bition (Goyarzu et al., 2004), and promote the expression of brainderived neurotrophic factor (BDNF) that is critical for neurogen-esis, also in adult animals, synaptic growth and neuron survival,especially in the learning- and memory-related brain regions suchas the hippocampus and subventricular zone (Kim et al., 2006;Valente et al., 2009). Second, flavonoids facilitate production ofthe signaling molecule nitric oxide, which inhibits the incidenceof atheromatous plaque adhesion molecules causing inflammation(Gonzalez-Gallego et al., 2007), and importantly, improves vascu-lar endothelial function by relaxing the smooth muscle tissue ofblood vessels (e.g., Heiss et al., 2003; Schroeter et al., 2006). In thisway, flavanol-rich cocoa can impose vasodilation in a nitric oxide-

dependent way both at the cardiovascular and peripheral levels.This in turn results in enhanced cerebral blood flow and bloodperfusion throughout the central and peripheral nervous system(Fisher et al., 2003, 2006; Hollenberg et al., 2009), affording better

A.N. Sokolov et al. / Neuroscience and Biobeh

neurogenesis, synap�c growth, neuron survival

MAPK, ERK, PI3 signaling

gene expression,protein synthesis

expression of BDNF

LTP, memory forma�on

neurocogni�ve func�ons

Cocoa flavanols

vasodila�on (heart, CNS, sensory systems)

nitric oxide synthesis

inflamma�on

vascular endothelian func�on, angiogenesis

oxygen, glucose supply

neuroprotec�on

Fig. 1. Schematic of putative neurobiological actions of cocoa flavanols; small up-and downward inclined arrows show up and down regulation of function, respec-tively. The absorbed flavanols either directly interact with cellular cascades (left,green), yielding expression of neuroprotective and neuromodulatory proteins thatpromote neurogenesis, neuronal function and brain connectivity, or act by improv-ing central and peripheral blood flow and angiogenesis in the brain and sensorysystems (right, yellow). Note the two action pathways may cross-talk, contribut-ing to more than one function. For example, improved neurogenesis, synapticgrowth, and neuron survival may promote neuroprotection, while neurocognitionmay benefit from increased oxygen and glucose supply due to vasodilation. LTP,long-term potentiation; mitogen-activated protein (MAPK), extracellular-signal-regulated (ERK), and phosphoinositide 3-kinase (PI3-kinase/Akt) cascades; BDNF,bow

smtma

flataaevaeflnna

3n

(aa(d

rain derived neurotophic factor; CNS, central nervous system. (For interpretationf the references to color in the figures throughout the article, please consult theeb version of this article.)

upply of oxygen and glucose to the neurons and removal of wasteetabolites in the brain and sensory systems (e.g., blood delivery

o the retina; Huber et al., 2006; Kalt et al., 2010). In addition, ani-al models indicate that cocoa flavonoid administration stimulates

ngiogenesis in the hippocampus (van Praag et al., 2007).In a nutshell, the multiple neurobiological actions of cocoa

avonoids in enhancing cognitive function and behavior may bettributed to the expression of neuroprotective and neuromodula-ory proteins that increase the number and connectivity of neurons,nd improve neuronal function. The other class of putative mech-nisms is related to the effects on vascular function that throughnhanced blood perfusion both in the brain and peripheral ner-ous system including sensory systems, may improve neuro- andngiogenesis (Williams and Spencer, 2012), cognitive function, andxecutive control. However, while the long-term actions of cocoaavanols in counteracting oxidative stress, neuroinflammation andeurodegeneration appear to be better understood, the mecha-isms of more immediate neuromodulatory effects on cognitionnd behavior remain unclear.

. Neuroprotective action of cocoa flavanols in aging andeurological disease

A number of health conditions in typical and atypical agingAlzheimer’s disease, vascular dementia, Parkinson’s disease) and

cute neurological conditions such as stroke have been associ-ted with disturbances of cerebral blood flow and oxidative stressMcGeer and McGeer, 2003; Hirsch et al., 2005). Potent antioxi-ant action and endothelium-dependent vasodilation capacity of

avioral Reviews 37 (2013) 2445–2453 2447

flavonoids may effectively protect against neural and oxidativedamage eventually giving rise to neurological disease, cognitive andfunctional decline.

3.1. Animal studies

Several animal models of normal and pathological aging exam-ined the effects of high-flavanol cocoa on the onset of age-relatedcognitive deficits. Rozan et al. (2007) assessed the preventive effectsof Acticoa powder, a cocoa polyphenolic extract, on free radi-cal production by leucocytes in rats following heat exposure andprotective effects of the cocoa powder on subsequent cognitive per-formance. Either high-flavanol cocoa powder (22.9 mg/kg/day) orvitamin E, as the antioxidant reference, was administered orally torats for 14 days prior to heat exposure. The day after heat expo-sure, free radical production in rats treated with cocoa powder orvitamin E was significantly reduced compared to controls. Unlikecontrols, cocoa powder and vitamin E-treated rats discriminatedbetween active and inactive levers in a light extinction paradigm.Throughout testing, the treated rats also exhibited decreasedescape latencies for reaching the hidden platform in the Morriswater maze. The results suggest that the daily oral administrationof Acticoa powder or vitamin E counteract the overproduction offree radicals, and thereby protect rats from cognitive impairmentscaused by heat exposure.

Bisson et al. (2008) administered orally Acticoa powder torats (as they grew from 15 to 27 months of age) at the dose of24 mg/kg/day. In aged rats, high-flavanol cocoa improved cogni-tive performance in light extinction and water maze paradigms,increased lifespan and preserved high urinary free dopamine lev-els. Acticoa powder appears to retard age-related impairments,including cognitive decline in normal aging and neurodegenerativediseases.

Converging evidence comes from animal models of neurode-generative diseases (such as Alzheimer’s disease), which make useof polyphenol-enriched diets. Fernández-Fernández et al. (2012)examined an ability of a diet rich in polyunsaturated fatty acidsand polyphenols from dry fruits and cocoa (called LMN diet, apatented product known to induce hippocampal neurogenesis inadult mice) for counteracting the age-related impairment and neu-ropathology in wild type and transgenic mice (Tg2576 genotypewith over-expression of the human APP gene carrying the Swedishmutation, K670N:M671L), an animal model of Alzheimer’s disease.At the age of 13 months, once the amyloid plaques (A�) wereformed, both mice types received LMN diet for further five months.In the last two months, they performed on a behavioral test bat-tery. Overall, both typically aging wild and, to a greater degree,Tg mice exhibited reduction in sensorimotor reflexes, exploratorybehavior in the hole board, activity in the elevated plus-maze,ambulation in the home cage during the dark phase, and in spatiallearning in the Morris water maze. The diet did not impact the detri-mental effects observed in sensorimotor reflexes, but did clearlyreverse the behavioral effects of both aging and Tg genotype. Thisbehavioral improvement correlated with a 70% increase in cellularproliferation in the subventricular zone of the brain, rather thanwith a decrease of amyloid plaques. In contrast, LMN diet admin-istered at an age of 10 months (before the plaques occurred) ledto a decreasing tendency of soluble and fibrillar A� levels in thehippocampus along with a decrease in A� plasma content, sug-gesting a putative role of the diet in delaying plaque formation.This is the first evidence that LMN diet can prevent behavioraldeterioration caused by aging and Tg genotype, and delay the

A� plaque formation. The results also highlight the increasingimportance of polyphenols from dry fruits and cocoa as humandietary supplements in amelioration of the cognitive and functionalimpairment during aging and neurological disease. Arendash et al.

2 iobeh

(aabawTpde

ompvthrtvpsact

3

acgeeas1EnaflucapKoninmmabw∼Tfss

(upwn

448 A.N. Sokolov et al. / Neuroscience and B

2007), however, administered a polyphenol-rich omega-3 fattycid diet of Greenland Eskimos to another type of Tg mice (anothernimal model of Alzheimer’s disease), and did not find any dietenefits, except for several behavioral measures such as open fieldctivity and maze entries. In both studies, Tg mice were similar,ith Arendash et al. (2007) using a second-generation cross of the

g2576 and the 6.2 lines that carried an additional mutation in theresenilin gene 1, PS1. The distinct outcomes were likely due to theifference in either diets or study protocols to which the mice werexposed.

In an animal model of stroke, Shah et al. (2010) administeredrally mice with 5, 15, or 30 mg/kg epicatechin 90 min beforeiddle cerebral artery occlusion. Epicatechin-treated mice, com-

ared to the control group, exhibited significantly smaller lesionolumes and improved neurological scores. Mice that were post-reated with 30 mg/kg of epicatechin at 3.5 h after the occlusionad also significantly smaller infarct volumes and improved neu-ological scores. Villarreal-Calderon et al. (2010) reported that areatment with dark chocolate prevents the inflammation of theagus nerve resulting from a 16-month exposure of mice to theolluted air of Mexico City. Mice exposed to polluted air had aignificant imbalance in genes coding for antioxidant defenses,poptosis and neurodegeneration at the level of the dorsal vagalomplex and this imbalance was mitigated by chocolate adminis-ration.

.2. Human population studies

Observational human population studies on neuroprotectivend neuromodulatory action of flavonoids, including high-flavanolocoa, have often been poorly controlled, whereas prospective lon-itudinal studies remain laborious and costly. To date, therefore,vidence for beneficial neuroprotective and anti-inflammatoryffects of cocoa flavanols on cognitive and behavioral decline inging and neurological disease is rather limited. In a prospectivetudy of a large sample of men aged 69–89 years (Kalmijn et al.,997), risk of cognitive decline (assessed by the Mini-Mental Statexamination) tended to inversely relate to flavonoid intake, thougho association was found between the risk of cognitive declinend vitamin C- or E-intake served as the antioxidant reference toavonoids. In a cross-sectional study of the elderly Norwegian pop-lation, Nurk et al. (2009) investigated the association betweenognitive performance and flavonoid intake from chocolate, wine,nd tea. Participants aged 70–74 years (n = 2031; 55% females) com-leted a comprehensive cognitive test battery consisting of theendrick Object Learning Test, Trail Making Test (part A), versionsf the Digit Symbol Test, Block Design, Mini-Mental State Exami-ation, and Controlled Oral Word Association Test. Habitual food

ntake was assessed by a self-reported food frequency question-aire. Chocolate, wine, or tea consumers yielded significantly betterean test scores and lower prevalence of poor cognitive perfor-ance. Those consuming all three items had the best test scores

nd the lowest risk for poor test performance. The associationetween intake of the foods and cognition were dose-dependent,ith sharp improvements at intakes of ∼10 g/day chocolate and75–100 ml/day wine, and a linear improvement for tea intake.he effect was most pronounced for wine and modestly weakeror chocolate intake. It appears that in the elderly, a diet high inome flavonoid-rich foods is associated with better performance ineveral cognitive abilities in a dose-dependent manner.

Participants of a prospective 10-year Personnes Agées QuidPAQUID) study (n = 1640; aged ≥65 years, free from dementia)

nderwent testing of cognitive function at four consecutive timeoints (Letenneur et al., 2007). At each visit, the cognitive functionsere assessed by test battery including Mini-Mental State Exami-ation, Benton’s Visual Retention Test, and “Isaacs” Set. Flavonoid

avioral Reviews 37 (2013) 2445–2453

intake from multiple food items including chocolate (assessed onceas the study begun) was associated with both better cognitiveperformance at baseline and better evolution of performance overtime. The most positive evolution was found in participants inthe two highest quartiles of flavonoid intake compared to thosein the lowest quartile. After 10-year follow-up, participants withthe lowest flavonoid intake had lost on average 2.1 points on theMini-Mental State Examination, compared to the 1.2-point loss inparticipants of the highest quartile. The study raises the possibilityof an association between dietary flavonoid intake and cognitiveaging.

3.3. Clinical studies

The known benefits of flavonoids for vascular health (Ried et al.,2012) may represent a promising approach in treating cerebrovas-cular disorders and protecting cognitive and functional behaviorin the elderly. Age- and disease-related disturbances in cerebralblood flow are thought to be commonly accompanied by cognitiveand behavioral decline. Dietary high-flavanol cocoa intake is asso-ciated with an increased cerebral blood flow velocity in the middlecerebral artery, suggesting a promising role for high-flavanol cocoaconsumption in the treatment of cerebrovascular ischemic syn-dromes such as dementia and stroke. For instance, Sorond et al.(2008) report that following two weeks of high-flavanol cocoaintake, in 34 healthy elderly volunteers (aged 72 ± 6 years; 16males), mean blood flow velocity in the middle cerebral artery mea-sured by transcranial Doppler ultrasound significantly increases by8% at one week and 10% at two weeks.

Potent antioxidative and anti-inflammatory properties offlavonoids have been proposed to play a role in preventing mildcognitive impairment, a precursor of dementia, and Alzheimer’sdisease. In Alzheimer’s disease, an increased production and accre-tion of A�-peptides activates microglia, resulting in release ofinflammatory mediators that further enhance A� production, giv-ing rise to neuronal dysfunction and cellular death. While �-and �-secretase facilitate A� production, �-secretase inhibits it.Recent work in cultured human neuroblastoma cells shows thatlow concentrations of nitric oxide up-regulate the expression of �-secretase and down-regulate that of �-secretase. This suggests thecerebrovascular nitric oxide might inhibit A� production (McCarty,2006; Pak et al., 2005). Cocoa flavanols, especially epicatechin, actdirectly on the endothelium of brain vessels, stimulating activity ofthe endothelial nitric oxide synthase that in turn induces vasodila-tion and improves cerebrovascular perfusion (Fisher et al., 2006;Schroeter et al., 2006; Patel et al., 2008). So far, there does notseem to be any proven association between intake of antioxidantsand vitamins and Alzheimer’s disease (see Luchsinger and Mayeux,2004 for review), however, several studies did report a diminishedcerebral blood flow in dementia patients (Ruitenberg et al., 2005;Nagahama et al., 2003). Cerebrovascular atrophy is also known tolead to mild cognitive impairment and subsequently to Alzheimer’sdisease. It is therefore conceivable that beneficial properties ofcocoa flavanols may slow down the transition from mild cogni-tive impairment to Alzheimer’s disease (Nagahama et al., 2003).Commenges et al. (2000) conducted a clinical trial with 1367 par-ticipants aged over 65 years, 66 from which developed dementia.The relative risk of developing dementia adjusted for age for the twohighest consumptions of flavonoids was 0.55 (95% confidence inter-val, 0.34–0.90). After further adjustment for sex, education level,weight and vitamin C intake, the relative risk decreased to 0.49 (95%

confidence interval, 0.26–0.92). Thus, it appears that antioxidantflavonoid intake is inversely related to the risk of dementia. How-ever, in this study flavonoids came mainly from fruits, vegetables,wine and tea rather than cocoa.

iobeh

uw∼oMflEhtlcgavow

fsl2irsat(caMaDtosfidwht

tiiiepdHt2mSodt

iicietmo

A.N. Sokolov et al. / Neuroscience and B

Most recently, Desideri et al. (2012) studied 90 elderly individ-als with mild cognitive impairment (mean age, 71 years; 43 males)ho consumed once daily for eight weeks a drink containing either990 mg (high), ∼520 mg (moderate), or ∼45 mg (low-flavanol)f cocoa flavanols. Cognitive function was assessed by the Mini-ental State Examination, Trail Making Test A and B, and verbal

uency test. At the end of the follow-up period, Mini-Mental Statexamination scores were similar in the treatment groups. In theigh- and moderate-flavanol compared to the low-flavanol groups,ime required to complete both Trail Making Tests was significantlyower. Verbal fluency score was significantly better in the high-ompared to low-flavanol group. The high- and moderate-flavanolroups also exhibited decreased insulin resistance, blood pressure,nd lipid peroxidation. This appears to be the first dietary inter-ention study to demonstrate the efficacy of regular consumptionf cocoa flavanols for improving cognitive function in the elderlyith mild cognitive impairment.

The ability of flavonoids to improve and maintain vascularunction offers a further possibility to investigate the relation-hip between cocoa flavanol intake and neuronal and functionaloss after stroke (for recent animal data, see, e.g., Shah et al.,010; Section 3.1). In the meta-analysis of three human stud-

es comprising 114,009 participants, Buitrago-Lopez et al. (2011)eported a 29% reduction of stroke risk in high chocolate con-umers compared to low consumers. Buijsse et al. (2010) foundn even stronger inverse correlation between chocolate consump-ion and stroke risk than for myocardial infarction. Rautiainen et al.2012) examined the association between the total antioxidantapacity (including fruits, vegetables, tea, coffee, and chocolate)nd stroke risk in women, aged 49 to 83 years, from the Swedishammography cohort. The study included 5680 women with

nd 31,035 women without a history of cardiovascular disease.iet was assessed with a self-reported food frequency ques-

ionnaire, and dietary total antioxidant capacity calculated usingxygen radical absorbance capacity values. Stroke cases wereubdivided into cerebral infarctions, hemorrhagic and unspeci-ed strokes. Using multivariate analyses with hazard ratios, theietary antioxidant capacity was found to be inversely associatedith total stroke in disease-free women (17% risk reduction) andemorrhagic stroke in women with disease history (45% risk reduc-ion).

In Parkinson’s disease, abnormal action of the neuromodula-or adenosine, which fails to suppress unwanted motor activityn the basal ganglia via striatopallidal neurons, has been linked tompaired motor function (Jankovic, 2008). Adenosine antagonists,ncluding caffeine and chocolate, have therefore been consid-red for ameliorating parkinsonian motor dysfunction. Parkinson’satients do report an increased chocolate consumption, indepen-ent of concomitant depressive symptoms (Wolz et al., 2009).owever, in an investigator-blinded, placebo-controlled, crossover

rial in 26 patients with moderate Parkinson’s disease (Wolz et al.,012), a single acute dose of dark chocolate failed to improveotor function (assessed by Unified Parkinson’s Disease Rating

cale motor score) over white flavanol-free white chocolate. Theutcome might likely be due to lacking patient blindfolding, theose of chocolate used, its flavanol content, and the time frame ofreatment.

To the most part, the neuroprotective potential described aboves attributable to cocoa and chocolate flavanols rather than otherngredients such as caffeine that has been widely implicated inounteracting age- and disease-related cognitive decline such asn Alzheimer’s and Parkinson’s disease (Costa et al., 2010; Santos

t al., 2010). Unlike in coffee, tea and soft drinks, the caffeine (andheobromine) concentration in chocolate and cocoa products is

uch lower than that of flavanols to account for potential effectsf chocolate (Benton, 2004).

avioral Reviews 37 (2013) 2445–2453 2449

4. Neuromodulation of cognition, mood, learning, andmemory

4.1. Animal studies

Flavonoids are believed to trigger expression of neuromodula-tory proteins in the brain regions implicated in learning, memory,and cognition, suggesting cocoa flavanols can exhibit immediateand short-term action on neurocognition, mood, and behavior. Sur-prisingly, only a few animal studies have addressed these issues.Mice treated with one of the major chocolate flavanols, epicat-echin, at the dose of 500 �g/g (daily supply of 2.5 mg) showedpronounced angiogenesis in the hippocampus (van Praag et al.,2007). Epicatechin treatment combined with exercise (runninga wheel) improved retention of spatial memory and increaseddendritic spine density in the dentate gyrus of the hippocam-pus. Moreover, epicatechin treatment facilitated gene expressionassociated with learning in the hippocampus but did not affect hip-pocampal adult neurogenesis. Yamada et al. (2009) compared theeffects of short-term versus long-term (two-week) oral administra-tion of cocoa mass in large amounts (100 mg/100 g body weight) inthe rat elevated T-maze test, an animal model of anxiety. Short-term administration significantly abolished avoidance behaviorduring immediate test performance, suggesting a reduced fear con-ditioning. Long-term administration enhanced brain concentrationof emotion-related neurotransmitter serotonin and its turnover.The findings indicate short-term cocoa intake shows an anxio-lytic effect, whereas long-term intake affects brain monoaminemetabolism. This suggests flavanol impacts on the amygdala under-pinning regulation of anxiety and encoding of affective valence (e.g.,Morrison and Salzman, 2010). Flavanol action therefore may occurin the brain regions outside the hippocampus and subventricularzone, in which it has already been established.

4.2. Human studies

In humans, several studies have aimed to identify immediateand short-term action of cocoa flavanols on mood and cognitiveperformance with as yet inconclusive outcome. Crews et al. (2008)had healthy older adults (n = 101; 41 males; age, ≥60 years) toconsume daily for 6 weeks either a 37-g bar of dark chocolateand 8 ounces (237 ml) of artificially sweetened cocoa beverage orsimilar placebo products. Participants underwent hematological,blood pressure, and pulse rate measurements, and accomplishedseveral cognitive tests: Selective Reminding, Wechsler MemoryScale-III Faces I and Faces II subtests, Trail Making Test, Stroop Test,Wechsler Adult Intelligence Scale-III Digit Symbol-Coding subtest,and General Activation subscale of the Activation-DeactivationAdjective Check List (A-DACL). The only effect observed was asignificantly higher, compared to the placebo group, pulse rate,with no effects found on blood pressure, hematological, and cogni-tive variables. In a randomized, double-blind, controlled, balanced,three period crossover study, 30 healthy young adults (meanage, 22 years; 13 males) consumed high-flavanol (520 mg and994 mg) cocoa drinks and a matched control drink, with a three-daywashout between drinks (Scholey et al., 2010). Over a 1 h testingperiod, participants repeatedly performed 10-min cycles of a Cogni-tive Demand Battery (two serial subtraction tasks, Serial Threes andSerial Sevens), a Rapid Visual Information Processing (RVIP) task,and a “mental fatigue” scale. High-flavanol cocoa intake improvedSerial Threes performance. The 994-mg beverage yielded speededRVIP responses, but also more errors during Serial Sevens. Only the

520-mg beverage attenuated self-reported “mental fatigue”. Thisis the first report on immediate improvements of cognitive func-tion following high-flavanol cocoa consumption in healthy youngadults.

2 iobehavioral Reviews 37 (2013) 2445–2453

yochbgteaef

eoA(1tRWWDm1nhaopfds(b

i((oad(as

ywrojscowtataod

cflNd

Fig. 2. Increased blood flow response (±SEM, standard error of mean) in the graymatter of four female participants following ingestion of an acute dose of highcompared to low flavanol drink.

From Francis et al., 2006. The effect of flavanol-rich cocoa on the fMRI response to a

450 A.N. Sokolov et al. / Neuroscience and B

In healthy young adults (n = 30, eight males; age range, 18–25ears) who in a crossover, order counterbalanced design consumednce either 35 g dark chocolate (with 720 mg of high-flavanolocoa) or a matched quantity of flavanol-free white chocolate,igh-flavanol cocoa improved visual contrast sensitivity (assessedy reading numbers varying in their luminance relative to back-round), working memory for location, choice reaction time, andhe time required to detect direction of coherent motion (Fieldt al., 2011). This outcome extends the range of cognitive functionsffected by high-flavanol cocoa consumption and provides the firstvidence on the immediate effects of high-flavanol cocoa on visualunctions.

In 72 healthy participants aged 40–65 years, Pase et al. (2013)xamined immediate and sub-chronic effects of cocoa flavanolsn mood and cognition. Three separate groups were assigned ancticoa dark chocolate drink mix containing 500, 250, or 0 mgplacebo) of polyphenols once daily for 30 days. At baseline, at, 2.5, and 4 h after a single acute dose, and again after 30-dayreatment, performance on cognitive tasks (Simple and Choiceeaction Time, Digit Vigilance, Tracking, Spatial and Numericorking Memory, Immediate and Delayed Word Recall, Delayedord and Picture Recognition) was assessed with the Cognitive

rug Research battery (Kennedy et al., 2002) and self-reportedood with the Bond-Lader Visual Analogue Scale (Bond and Lader,

974). While mood remained unaffected by the acute treatment,o effects on cognition occurred at all time points. At 30 days, theigh-dose treatment significantly improved self-rated calmnessnd contentedness compared to placebo. This is the first evidencen the effects of cocoa polyphenols on mood in healthy partici-ants. The outcome suggests a possibility for cocoa polyphenolsor ameliorating the symptoms associated with clinical anxiety andepression. Although some recent studies do indicate a relation-hip between chocolate consumption and depressive symptomse.g., Rose et al., 2010), the nature of the relationship remains toe clarified.

Latest developments in modern non-invasive neuroimag-ng techniques such as functional magnetic resonance imagingfMRI), electroencephalography, and magnetoencephalographyEEG/MEG), have made it possible to explore modulatory effectsf food constituents on neural activity in the human brain and itsssociation with behavior. Yet, with the exception of some workone on the modulation of brain reward system by chocolate intakee.g., Small et al., 2001; Stice et al., 2008; Valentin et al., 2007), only

few neuroimaging studies have examined the neural networkselectively activated by cocoa flavanols.

Francis et al. (2006) used a paired letter-digit task in healthyoung female participants (n = 16; age range, 18–30 years)hile recording the blood oxygenation level-dependent (BOLD)

esponses in an fMRI protocol following 5-day ingestion of 150 mgf cocoa flavanols. Participants had to press a key, making either oneudgment in the “no switch” condition (“is the letter a vowel or con-onant”, “is the digit even or odd”) or two judgments in the “switch”ondition (responding both to the letter and digit). Although anverall BOLD signal increased during the cognitive task, no effectsere found in response time, switch cost (the difference in response

ime between “switch” and “no-switch” conditions), and heart ratefter consumption of this moderate dose of cocoa flavanols. Inges-ion of a single acute dose (450 mg of cocoa flavanols) yieldedn increased cerebral blood flow (Fig. 2), confirming a potentialf cocoa flavanols for treatment of vascular impairment such asementia and stroke (see Section 3).

Over a 30-day period, Camfield et al. (2012) administered a daily

hocolate drink (250 mg or 500 mg cocoa flavanols versus low-avanol cocoa as placebo) to 63 volunteers (aged 40–65 years).eurocognitive changes associated with flavanol supplementationuring performance of a spatial working memory task at baseline

cognitive task in healthy young people. Journal of Cardiovascular Pharmacology 47,Suppl. 2, S215–20. Copyright © 2006 Lippincott Williams & Wilkins with permissionof Wolters Kluwer Health/Lippincott Williams & Wilkins.

and at the end of the treatment were assessed using EEG (steadystate visually evoked potentials, SSVP). Behavioral measures ofaccuracy and response time did not differ in a dose-dependentmanner, whereas average amplitude and phase of the evokedpotentials at a number of posterior parietal and centro-frontal sitessignificantly differed between the groups during memory encoding,the memory hold period, and retrieval (Fig. 3). The authors assumethe differences in brain activation, even in the absence of behavioraleffects, point to an increased neural efficiency in spatial workingmemory associated with chronic cocoa flavanol consumption.

Table 1 summarizes the results of research on the neuroprotec-tive action of flavanols in aging and neurological diseases on theone hand, and the neuromodulatory effects on cognition, mood,learning, and memory on the other hand. As seen from the table,basic animal intervention studies on mechanisms of action of fla-vanols still outnumber human and clinical studies demonstratingthat these effects are of physiological and clinical relevance.

In summary, recent research has yielded encouraging albeit asyet inconclusive outcome. For exploring possible effects of high-flavanol cocoa on human behavior, cognitive and brain functions,standardized psychometric tasks and neuroimaging protocols arerequired along with administering proper high-flavanol cocoadosages. To date, there is still limited evidence for high-flavanolcocoa effects on human cognitive and affective function, and behav-ior. Moreover, it remains unclear how high-flavanol cocoa modu-lates the brain networks underlying neural cognitive processing.

5. Concluding remarks and future directions

Flavonoids, potent antioxidant and anti-inflammatory agents,represent up to 20% of compounds found in cocoa beans. Fla-vanols, and especially epicatechin, are the most common cocoaflavonoids. The flavanol contents in cocoa products and chocolatevary greatly depending on the bean variety and origin, agricul-tural and processing practices. In part, the variability of flavanolcontents in cocoa and chocolate may be responsible for the mixedoutcomes presently observed in research on the effects of cocoa

flavanols on neurocognitive and affective function, executive con-trol, and behavior. This might be due to the fact that to date, onlyfew – especially, human – intervention studies have directly exam-ined exposure to cocoa flavonoids, while the most data comes from

A.N. Sokolov et al. / Neuroscience and Biobehavioral Reviews 37 (2013) 2445–2453 2451

Fig. 3. Topographic differences in the average steady state visually evoked potentials SSVEP amplitudes at baseline and retest for a spatial working memory task (encoding,hold interval, and retrieval) in the low, medium and high cocoa flavanol (CF) groups. Warm and cool colors show SSVP amplitude decreases and increases, respectively,post-treatment compared to baseline. (For interpretation of the references to color in the figures throughout the article, please consult the web version of this article.)

From Camfield et al., 2012. Steady state visually evoked potential (SSVEP) topography changes associated with cocoa flavanol consumption. Physiology and Behavior 105,948–57. Copyright © 2011 Elsevier Inc. with permission of Elsevier Ltd.

Table 1Cocoa flavonoid effects on the brain, cognition, and behavior – overview.

Brain region/function Animal models Human studies

Whole brainImproved cerebral blood flow (general

improvement of neural function dueto enhanced oxygen and glucosedelivery)

15 8, 9, 10, 19, 24, 25

HippocampusNeurogenesis 15, 26 28Improved neuronal connectivity 27Angiogenesis 27

Subventricular zone of the brainNeurogenesis 6, 26

VisionImproved blood delivery to the retina 14 11Rapid visual processing 23Visual contrast sensitivity 7Coherent motion direction 7

PerformanceChoice reaction time 7

Cognition, learning, and memory; moodGeneral 22 16Mental calculation 23Working memory for location 7Fear conditioning (anxiety) 29Calmness, contentedness 20

AgingReduced cognitive decline 1, 6 13, 16, 18, 25

Neurological diseaseAlzheimer (reduced cognitive decline) 4, 5Stroke (reduced risk) 2, 3, 12, 17, 21

Key to references: 1: Bisson et al. (2008), 2: Buijsse et al. (2010), 3: Buitrago-Lopezet al. (2011), 4: Commenges et al. (2000), 5: Desideri et al. (2012), 6: Fernández-Fernández et al. (2012), 7: Field et al. (2011), 8: Fisher et al. (2003), 9: Fisher et al.(2006), 10: Francis et al. (2006), 11: Huber et al. (2006), 12: Janszky et al. (2009), 13:Kalmijn et al. (1997), 14: Kalt et al. (2010), 15: Kim et al. (2006), 16: Letenneur et al.(2007), 17: Mink et al. (2007), 18: Nurk et al. (2009), 19: Patel et al. (2008), 20: Paseet al. (2013), 21: Rautiainen et al. (2012), 22: Rozan et al. (2007), 23: Scholey et al.(2010), 24: Schroeter et al. (2006), 25: Sorond et al. (2008), 26: Valente et al. (2009),27: van Praag et al. (2007), 28: Vauzour et al. (2008), 29: Yamada et al. (2009).

research using dark chocolate as flavanol supply. Despite multi-plicity of flavanol effects in the brain, neurobiological actions offlavanols are believed to occur in two major ways: (i) via directinteractions with cellular and molecular signaling cascades, espe-cially in the brain regions dedicated to learning and memory,and (ii) via central and peripheral blood-flow improvement andangiogenesis, including the brain and sensory systems. Overall, evi-dence on the persistence of neuroprotective and neuromodulatoryactions of cocoa flavanols, albeit as yet limited and inconclusive,suggests that cocoa flavanols may exert both long lasting andimmediate effects on neurocognition and behavior. The immedi-ate effects can be attained with a single acute or subchronic (forseveral weeks) administration of cocoa flavanols in appropriatedosages. The lasting effects likely require chronic intake of flavanol-rich cocoa products over an extended time frame. In both instances,more controlled follow-up studies should be in place to determinethe effects’ duration.

While long lasting neuroprotective properties of flavonoidintake for neurodegenerative and neuroinflammatory diseases ofthe nervous system have been relatively well documented (Katzet al., 2011; although more research specifically on cocoa flavanolsand greater, e.g. population-based, participant samples is required),more immediate action on cognitive and affective function, exec-utive control, and behavior remains largely unknown, as are sexdifferences in both long-term and immediate actions (see Sokolovet al., 2013 for a companion review). Sex-related variability in datais likely to account for the lack of significant effects of cocoa fla-vanols on the brain responses and cognitive performance observedin some of the studies. In females, distinct phases of the menstrualcycle and the perimenopause may also add to data variability.

The main issues to resolve in the studies to come include (i)blindfolding both the experimenter and participants, which isimportant for reducing the well-known reward value of chocolateintake, (ii) determining an appropriate (immediate vs. short-termvs. long-term) time frame and dose of flavanol administration, (iii)choice of simple tests of cognitive, executive and affective function

and performance that possess sufficient sensitivity and specificity(Macready et al., 2009), and (iv) for examining neuroprotectiveproperties of flavanol consumption, the onset and duration ofconsumption. Another important issue to consider is a so called

2 iobeh

fiFmctgFsncoSFc

nrim

A

FsBH

R

A

A

A

B

B

B

B

B

B

B

B

C

C

C

C

452 A.N. Sokolov et al. / Neuroscience and B

ood matrix, or food composition, in which cocoa flavonoids appearn food (e.g., Lamuela-Raventós et al., 2005; Manach et al., 2004).or example, ingestion of 100 g dark chocolate along with 200 mlilk results in a substantial reduction of both total antioxidant

apacity and (-)epicatechin content of human plasma, comparedo ingestion of 100 g pure dark chocolate, and the reduction is evenreater after ingestion of 200 g milk chocolate (Serafini et al., 2003).inally, other constituents of cocoa and plants rich in flavanols,uch as their relatively high content of tryptophan, a precursor ofeurotransmitter serotonin (e.g., Bertazzo et al., 2011), may eitherontribute to the neurobiological effects of flavanols or, dependingn the constituents’ bioavailability (e.g., due to the food matrix;mit, 2011), exert different actions through different pathways.orthcoming studies should explore whether these actions areomplementary, antagonistic or synergistic.

In conclusion, future research has to combine functionaleuroimaging techniques such as fMRI, EEG and MEG with neu-ocognitive and behavioral correlates to uncover long lasting andmmediate effects of chocolate consumption on human cognition,

ood, and behavior.

cknowledgements

We thank Otmar and Edelgard Knoll and the Willy Robert Pitzeroundation for support. MAP was supported by the Else Kröner Fre-enius Foundation (Grants P2010 92 and P2013 127), the Reinholdeitlich Foundation, the Berthold Leibinger Foundation, and by theeidehof Foundation.

eferences

ndrés-Lacueva, C., Shukitt-Hale, B., Galli, R.L., Jauregui, O., Lamuela-Raventós, R.M.,Joseph, J.A., 2005. Anthocyanins in aged blueberry-fed rats are found centrallyand may enhance memory. Nutr. Neurosci. 8, 111–120.

rendash, G.W., Jensen, M.T., Salem Jr., N., Hussein, N., Cracchiolo, J., Dickson, A.,Leighty, R., Potter, H., 2007. A diet high in omega-3 fatty acids does not improveor protect cognitive performance in Alzheimer’s transgenic mice. Neuroscience149, 286–302.

rts, I.C.W., Hollman, P.C.H., Kromhout, D., 1999. Chocolate as a source of teaflavonoids. Lancet 354, 488.

enton, D., 2004. The biology and psychology of chocolate craving. In: Nehlig,A. (Ed.), Coffee, Tea, Chocolate and the Brain. CRC Press, Strasbourg,pp. 205–218.

ertazzo, A., Comai, S., Brunato, I., Zancato, M., Costa, C.V.L., 2011. The content of pro-tein and non-protein (free and protein-bound) tryptophan in Theobroma cacaobeans. Food Chem. 124, 93–96.

irks, J., Grimley Evans, J., 2009. Ginkgo biloba for cognitive impairment and demen-tia. Cochrane Database Syst. Rev. 1, CD003120.

isson, J.F., Nejdi, A., Rozan, P., Hidalgo, S., Lalonde, R., Messaoudi, M., 2008. Effectsof long-term administration of a cocoa polyphenolic extract (Acticoa powder)on cognitive performances in aged rats. Br. J. Nutr. 100, 94–101.

ond, A., Lader, M., 1974. The use of analogue scales in rating subjective feelings. Br.J. Med. Psychol. 47, 211–218.

ors, W., Michel, C., 2002. Chemistry of the antioxidant effect of polyphenols. Ann.N.Y. Acad. Sci. 957, 57–69.

uijsse, B., Weikert, C., Drogan, D., Bergmann, M., Boeing, H., 2010. Chocolate con-sumption in relation to blood pressure and risk of cardiovascular disease inGerman adults. Eur. Heart J. 31, 1616–1623.

uitrago-Lopez, A., Sanderson, J., Johnson, L., Warnakula, S., Wood, A., Di Ange-lantonio, E., Franco, O.H., 2011. Chocolate consumption and cardiometabolicdisorders: systematic review and metaanalysis. Br. Med. J. 343, d4488.

amfield, D.A., Scholey, A., Pipingas, A., Silberstein, R., Kras, M., Nolidin, K., Wesnes,K., Pase, M., Stough, C., 2012. Steady state visually evoked potential (SSVEP)topography changes associated with cocoa flavanol consumption. Physiol.Behav. 105, 948–957.

ommenges, D., Scotet, V., Renaud, S., Jacqmin-Gadda, H., Barberger-Gateau, P., Dar-tigues, J.F., 2000. Intake of flavonoids and risk of dementia. Eur. J. Epidemiol. 16,357–363.

osta, J., Lunet, N., Santos, C., Santos, J., Vaz-Carneiro, A., 2010. Caffeine exposure andthe Parkinson’s disease: a systematic review and meta-analysis of observationalstudies. Alzheimers Dis. 20 (Suppl. 1), S221–S238.

rews Jr., W.D., Harrison, D.W., Wright, J.W., 2008. A double-blind, placebo-controlled, randomized trial of the effects of dark chocolate and cocoa onvariables associated with neuropsychological functioning and cardiovascularhealth: clinical findings from a sample of healthy, cognitively intact older adults.Am. J. Clin. Nutr. 87, 872–880.

avioral Reviews 37 (2013) 2445–2453

DeCarli, C., 2012. Cerebrovascular disease: assessing the brain as an end-organ ofvascular disease. Nat. Rev. Cardiol. 9, 435–436.

Desideri, G., Kwik-Uribe, C., Grassi, D., Necozione, S., Ghiadoni, L., Mastroiacovo, D.,Raffaele, A., Ferri, L., Bocale, R., Lechiara, M.C., Marini, C., Ferri, C., 2012. Benefits incognitive function, blood pressure, and insulin resistance through cocoa flavanolconsumption in elderly subjects with mild cognitive impairment: the Cocoa,Cognition, and Aging (CoCoA) study. Hypertension 60, 794–801.

EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA), 2012. ScientificOpinion on the substantiation of a health claim related to cocoa flavanols andmaintenance of normal endothelium-dependent vasodilation pursuant to Arti-cle 13(5) of Regulation (EC) No 1924/2006. EFSA J. 10, 2809.

Engler, M.B., Engler, M.M., Chen, C.Y., Malloy, M.J., Browne, A., Chiu, E.Y., Kwak, H.K.,Milbury, P., Paul, S.M., Blumberg, J., Mietus-Snyder, M.L., 2004. Flavonoid-richdark chocolate improves endothelial function and increases plasma epicatechinconcentrations in healthy adults. J. Am. Coll. Nutr. 23, 197–204.

Fernández-Fernández, L., Comes, G., Bolea, I., Valente, T., Ruiz, J., Murtra, P., Ramirez,B., Anglés, N., Reguant, J., Morelló, J.R., Boada, M., Hidalgo, J., Escorihuela, R.M.,Unzeta, M., 2012. LMN diet, rich in polyphenols and polyunsaturated fatty acids,improves mouse cognitive decline associated with aging and Alzheimer’s dis-ease. Behav. Brain Res. 228, 261–271.

Field, D.T., Williams, C.M., Butler, L.T., 2011. Consumption of cocoa flavanols resultsin an acute improvement in visual and cognitive functions. Physiol. Behav. 103,255–260.

Fisher, N.D., Hughes, M., Gerhard-Herman, M., Hollenberg, N.K., 2003. Flavanol-rich cocoa induces nitric oxide-dependent vasodilation in healthy humans. J.Hypertens. 21, 2281–2286.

Fisher, N.D., Sorond, F.A., Hollenberg, N.K., 2006. Cocoa flavanols and brain perfusion.J. Cardiovasc. Pharmacol. 47 (Suppl. 2), S210–S214.

Francis, S.T., Head, K., Morris, P.G., Macdonald, I.A., 2006. The effect of flavanol-rich cocoa on the fMRI response to a cognitive task in healthy young people. J.Cardiovasc. Pharmacol. 47 (Suppl 2), S215–S220.

Gonzalez-Gallego, J., Sanchez-Campos, S., Tunon, M.J., 2007. Antiinflammatory prop-erties of dietary flavonoids. Nutr. Hospital. 22, 287–293.

Goyarzu, P., Malin, D.H., Lau, F.C., Taglialatela, G., Moon, W.D., Jennings, R., Moy, E.,Moy, D., Lippold, S., Shukitt-Hale, B., Joseph, J.A., 2004. Blueberry supplementeddiet: effects on object recognition memory and nuclear factor-kappa B levels inaged rats. Nutr. Neurosci. 7, 75–83.

Grassi, D., Desideri, G., Necozione, S., Lippi, C., Casale, R., Properzi, G., Blumberg,J.B., Ferri, C., 2008. Blood pressure is reduced and insulin sensitivity increasedin glucose-intolerant, hypertensive subjects after 15 days of consuming high-polyphenol dark chocolate. J. Nutr. 138, 1671–1676.

Grassi, D., Necozione, S., Lippi, C., Croce, G., Valeri, L., Pasqualetti, P., Desideri,G., Blumberg, J.B., Ferri, C., 2005. Cocoa reduces blood pressure and insulinresistance and improves endothelium-dependent vasodilation in hypertensives.Hypertension 46, 398–405.

Heiss, C., Dejam, A., Kleinbongard, P., Schewe, T., Sies, H., Kelm, M., 2003. Vasculareffects of cocoa rich in flavan-3-ols. J. Am. Med. Assoc. 290, 1030–1031.

Hirsch, E.C., Hunot, S., Hartmann, A., 2005. Neuroinflammatory processes in Parkin-son’s disease. Parkinson. Relat. Disord. 11 (Suppl. 1), S9–S15.

Hollenberg, N.K., Fisher, N.D., McCullough, M.L., 2009. Flavanols, the Kuna, cocoaconsumption, and nitric oxide. J. Am. Soc. Hypertens. 3, 105–112.

Hooper, L., Kay, C., Abdelhamid, A., Kroon, P.A., Cohn, J.S., Rimm, E.B., Cassidy, A.,2012. Effects of chocolate, cocoa, and flavan-3-ols on cardiovascular health: asystematic review and meta-analysis of randomized trials. Am. J. Clin. Nutr. 95,740–751.

Huber, K.K., Adams, H., Remky, A., Arend, K.O., 2006. Retrobulbar haemodynamicsand contrast sensitivity improvements after CO2 breathing. Acta Ophthalmol.Scand. 84, 481–487.

Hurst, W.J., Tarka Jr., S.M., Powis, T.G., Valdez Jr., F., Hester, T.R., 2002. Cacao usageby the earliest Maya civilization. Nature 418, 289–290.

Jankovic, J., 2008. Are adenosine antagonists, such as istradefylline, caffeine, andchocolate, useful in the treatment of Parkinson’s disease? Ann. Neurol. 63,267–269.

Janszky, I., Mukamal, K.J., Ljung, R., Ahnve, S., Ahlbom, A., Hallqvist, J., 2009.Chocolate consumption and mortality following a first acute myocardialinfarction: the Stockholm Heart Epidemiology Program. J. Intern. Med. 266,248–257.

Kalmijn, S., Feskens, E.J., Launer, L.J., Kromhout, D., 1997. Polyunsaturated fatty acids,antioxidants, and cognitive function in very old men. Am. J. Epidemiol. 145,33–41.

Kalt, W., Hanneken, A., Milbury, P., Tremblay, F., 2010. Recent research on polyphe-nolics in vision and eye health. J. Agric. Food Chem. 58, 4001–4007.

Katz, D.L., Doughty, K., Ali, A., 2011. Cocoa and chocolate in human health anddisease. Antioxid. Redox Signal. 15, 2779–2811.

Kay, C.D., Kris-Etherton, P.M., West, S.G., 2006. Effects of antioxidant-rich foods onvascular reactivity: review of the clinical evidence. Curr. Atheroscler. Rep. 8,510–522.

Kelleher III, R.J., Govindarajan, A., Jung, H.Y., Kang, H., Tonegawa, S., 2004. Transla-tional control by MAPK signaling in long-term synaptic plasticity and memory.Cell 116, 467–479.

Kennedy, D.O., Scholey, A.B., Wesnes, K.A., 2002. Modulation of cognition and

mood following administration of single doses of Ginkgo biloba, ginseng, anda ginkgo/ginseng combination to healthy young adults. Physiol. Behav. 75,739–751.

Kim, D.H., Jeon, S.J., Son, K.H., Jung, J.W., Lee, S., Yoon, B.H., Choi, J.W., Cheong,J.H., Ko, K.H., Ryu, J.H., 2006. Effect of the flavonoid, oroxylin A, on transient

iobeh

L

L

L

M

M

M

M

M

M

M

M

N

N

N

N

N

N

P

P

P

P

P

R

R

R

R

R

R

S

S

S

A.N. Sokolov et al. / Neuroscience and B

cerebral hypoperfusion-induced memory impairment in mice. Pharmacol.Biochem. Behav. 85, 658–668.

amuela-Raventós, R.M., Romero-Pérez, A.I., Andrés-Lacueva, C., Tornero, A., 2005.Review: health effects of cocoa flavonoids. Food Sci. Technol. Int. 11, 159–176.

etenneur, L., Proust-Lima, C., Le Gouge, A., Dartigues, J.F., Barberger-Gateau, P., 2007.Flavonoid intake and cognitive decline over a 10-year period. Am. J. Epidemiol.165, 1364–1371.

uchsinger, J., Mayeux, R., 2004. Dietary factors and Alzheimer’s disease. LancetNeurol. 3, 579–587.

acready, A.L., Kennedy, O.B., Ellis, J.A., Williams, C.M., Spencer, J.P., Butler, L.T., 2009.Flavonoids and cognitive function: a review of human randomized controlledtrial studies and recommendations for future studies. Genes Nutr. 4, 227–242.

anach, C., Scalbert, A., Morand, C., Rémésy, C., Jiménez, L., 2004. Polyphenols: foodsources and bioavailability. Am. J. Clin. Nutr. 79, 727–747.

ann, G.E., Bonacasa, B., Ishii, T., Siow, R.C., 2009. Targeting the redox sensitiveNrf2-Keap1 defense pathway in cardiovascular disease: protection afforded bydietary isoflavones. Curr. Opin. Pharmacol. 9, 139–145.

cCarty, M.F., 2006. Toward prevention of Alzheimer’s disease – potential nutraceu-tical strategies for suppressing the production of amyloid beta peptides. Med.Hypotheses 67, 682–697.

cGeer, E.G., McGeer, P.L., 2003. Inflammatory processes in Alzheimer’s disease.Progress Neuro-Psychopharmacol. Biol. Psychiatry 27, 741–749.

esserli, F.H., 2012. Chocolate consumption, cognitive function, and Nobel laure-ates. N. Engl. J. Med. 367, 1562–1564.

ink, P.J., Scrafford, C.G., Barraj, L.M., Harnack, L., Hong, C.P., Nettleton, J.A., Jacobs Jr.,D.R., 2007. Flavonoid intake and cardiovascular disease mortality: a prospectivestudy in postmenopausal women. Am. J. Clin. Nutr. 85, 895–909.

orrison, S.E., Salzman, C.D., 2010. Re-valuing the amygdala. Curr. Opin. Neurobiol.20, 221–230.

agahama, Y., Nabatame, H., Okina, T., Yamauchi, H., Narita, M., Fujimoto, N.,Murakami, M., Fukuyama, H., Matsuda, M., 2003. Cerebral correlates of theprogression rate of the cognitive decline in probable Alzheimer’s disease. Eur.Neurol. 50, 1–9.

ehlig, A., 2010. Is caffeine a cognitive enhancer? J. Alzheimer’s Dis. 20 (Suppl. 1),S85–S94.

ehlig, A., 2013. The neuroprotective effects of cocoa flavanol and its influence oncognitive performance. Br. J. Clin. Pharmacol. 75, 716–727.

eveu, V., Pérez-Jiménez, J., Vos, F., Crespy, V., du Chaffaut, L., Mennen, L., Knox, C.,Eisner, R., Cruz, J., Wishart, D., Scalbert, A., 2010. Phenol-Explorer: an onlinecomprehensive database on polyphenol contents in foods. Database (Oxf.),http://dx.doi.org/10.1093/database/bap024.

iemenak, N., Rohsius, C., Elwers, S., Ndoumou, D.O., Lieberei, R., 2006. Comparativestudy of different cocoa (Theobroma cacao L.) clones in terms of their phenolicand anthocyanins contents. J. Food Compos. Anal. 19, 612–619.

urk, E., Refsum, H., Drevon, C.A., Tell, G.S., Nygaard, H.A., Engedal, K., Smith, A.D.,2009. Intake of flavonoid-rich wine, tea, and chocolate by elderly men andwomen is associated with better cognitive test performance. Nutrition 139,120–127.

ak, T., Cadet, P., Mantione, K.J., Stefano, G.B., 2005. Morphine via nitric oxide modu-lates beta-amyloid metabolism: a novel protective mechanism for Alzheimer’sdisease. Med. Sci. Monitor 11, BR357–BR366.

ase, M.P., Scholey, A.B., Pipingas, A., Kras, M., Nolidin, K., Gibbs, A., Wesnes, K.,Stough, C., 2013. Cocoa polyphenols enhance positive mood states but not cog-nitive performance: a randomized, placebo-controlled trial. J. Psychopharmacol.27, 451–458.

assamonti, S., Vrhovsek, U., Vanzo, A., Mattivi, F., 2005. Fast access of some grapepigments to the brain. J. Agric. Food Chem. 53, 7029–7034.

atel, A.K., Rogers, J.T., Huang, X., 2008. Flavanols, mild cognitive impairment, andAlzheimer’s dementia. Int. J. Clin. Exp. Med. 1, 181–191.

érez-Jiménez, J., Fezeu, L., Touvier, M., Arnault, N., Manach, C., Hercberg, S., Galan,P., Scalbert, A., 2011. Dietary intake of 337 polyphenols in French adults. Am. J.Clin. Nutr. 93, 1220–1228.

autiainen, S., Larsson, S., Virtamo, J., Wolk, A., 2012. Total antioxidant capacity ofdiet and risk of stroke: a population-based prospective cohort of women. Stroke43, 335–340.

ichelle, M., Tavazzi, I., Enslen, M., Offord, E.A., 1999. Plasma kinetics in man ofepicatechin from black chocolate. Eur. J. Clin. Nutr. 53, 22–26.

ied, K., Sullivan, T.R., Fakler, P., Frank, O.R., Stocks, N.P., 2012. Effect of cocoa onblood pressure. Cochrane Database Syst. Rev. 8, CD008893.

ose, N., Koperski, S., Golomb, B.A., 2010. Mood food: chocolate and depressivesymptoms in a cross-sectional analysis. Arch. Intern. Med. 170, 699–703.

ozan, P., Hidalgo, S., Nejdi, A., Bisson, J.F., Lalonde, R., Messaoudi, M., 2007. Preven-tive antioxidant effects of cocoa polyphenolic extract on free radical productionand cognitive performances after heat exposure in Wistar rats. J. Food Sci. 72,S203–S206.

uitenberg, A., den Heijer, T., Bakker, S.L., van Swieten, J.C., Koudstaal, P.J., Hofman,A., Breteler, M.M., 2005. Cerebral hypoperfusion and clinical onset of dementia:the Rotterdam Study. Ann. Neurol. 57, 789–794.

antos, C., Costa, J., Santos, J., Vaz-Carneiro, A., Lunet, N., 2010. Caffeine intake anddementia: systematic review and meta-analysis,. J. Alzheimer Dis. 20 (Suppl. 1),S187–S204.

calbert, A., Andres-Lacueva, C., Arita, M., Kroon, P., Manach, C., Urpi-Sarda,M., Wishart, D., 2011. Databases on food phytochemicals and their health-promoting effects. J. Agric. Food Chem. 59, 4331–4348.

choley, A.B., French, S.J., Morris, P.J., Kennedy, D.O., Milne, A.L., Haskell, C.F., 2010.Consumption of cocoa flavanols results in acute improvements in mood and

avioral Reviews 37 (2013) 2445–2453 2453

cognitive performance during sustained mental effort. J. Psychopharmacol.(Oxf.) 24, 1505–1514.

Schroeter, H., Heiss, C., Balzer, J., Kleinbongard, P., Keen, C.L., Hollenberg, N.K., Sies,H., Kwik-Uribe, C., Schmitz, H.H., Kelm, M., 2006. (−)-Epicatechin mediates ben-eficial effects of flavanol-rich cocoa on vascular function in humans. Proc. Natl.Acad. Sci. U.S.A. 103, 1024–1029.

Serafini, M., Bugianesi, R., Maiani, G., Valtuena, S., De Santis, S., Crozier, A., 2003.Plasma antioxidants from chocolate. Nature 424, 1013.

Shah, Z.A., Li, R.C., Ahmad, A.S., Kensler, T.W., Yamamoto, M., Biswal, S., Doré, S., 2010.The flavanol (−)-epicatechin prevents stroke damage through the Nrf2/HO1pathway. J. Cereb. Blood Flow Metab. 30, 1951–1961.

Sies, H., Hollman, P.C., Grune, T., Stahl, W., Biesalski, H.K., Williamson, G., 2012.Protection by flavanol-rich foods against vascular dysfunction and oxidativedamage: 27th Hohenheim Consensus Conference. Adv. Nutr. 3, 217–221.

Small, D.M., Zatorre, R.J., Dagher, A., Evans, A.C., Jones-Gotman, M., 2001. Changes inbrain activity related to eating chocolate: from pleasure to aversion. Brain 124,1720–1733.

Smit, H.J., Blackburn, R.J., 2005. Reinforcing effects of caffeine and theobromine asfound in chocolate. Psychopharmacology (Berl.) 181, 101–106.

Smit, H.J., Gaffan, E.A., Rogers, P.J., 2004. Methylxanthines are the psycho-pharmacologically active constituents of chocolate. Psychopharmacology (Berl.)176, 412–419.

Smit, H.J., Rogers, P.J., 2000. Effects of low doses of caffeine on cognitive performance,mood and thirst in low and higher caffeine consumers. Psychopharmacology(Berl.) 152, 167–173.

Smit, H.J., 2011. Theobromine and the pharmacology of cocoa. In: Fredholm, B.B.(Ed.), Methylxanthines, Handbook of Experimental Pharmacology 200. Springer,Berlin/Heidelberg, pp. 201–234.

Sokolov, A.N., Pavlova, M.A., Klosterhalfen, S., Enck, P., 2013. Sex differences in choco-late action and consumption (in preparation).

Sorond, F.A., Lipsitz, L.A., Hollenberg, N.K., Fisher, N.D.L., 2008. Cerebral blood flowresponse to flavanol-rich cocoa in healthy elderly humans. Neuropsychiatr. Dis.Treatment 4, 433–440.

Stevenson, D.E., Hurst, R.D., 2007. Polyphenolic phytochemicals – just antioxidantsor much more? Cell. Mol. Life Sci. 64, 2900–2916.

Stice, E., Spoor, S., Bohon, C., Veldhuizen, M.G., Small, D.M., 2008. Relation of rewardfrom food intake and anticipated food intake to obesity: a functional magneticresonance imaging study, Journal of. Abnorm. Psychol. 117, 924–935.

Valente, T., Hidalgo, J., Bolea, I., Ramirez, B., Anglés, N., Reguant, J., Morelló, J.R.,Gutiérrez, C., Boada, M., Unzeta, M., 2009. A diet enriched in polyphenols andpolyunsaturated fatty acids, LMN diet, induces neurogenesis in the subven-tricular zone and hippocampus of adult mouse brain. J. Alzheimer Dis. 18,849–865.

Valentin, V.V., Dickinson, A., O’Doherty, J.P., 2007. Determining the neural substratesof goal-directed learning in the human brain. J. Neurosci. 27, 4019–4026.

van Praag, H., Lucero, M.J., Yeo, G.W., Stecker, K., Heivand, N., Zhao, C., Yip, E.,Afanador, M., Schroeter, H., Hammerstone, J., Gage, F.H., 2007. Plant-derived fla-vanol (−)epicatechin enhances angiogenesis and retention of spatial memory inmice. J. Neurosci. 27, 5869–5878.

Vauzour, D., Vafeiadou, K., Rodriguez-Mateos, A., Rendeiro, C., Spencer, J.P., 2008. Theneuroprotective potential of flavonoids: a multiplicity of effects. Genes Nutr. 3,115–126.

Villarreal-Calderon, R., Torres-Jardón, R., Palacios-Moreno, J., Osnaya, N., Pérez-Guillé, B., Maronpot, R.R., Reed, W., Zhu, H., Calderón-Garciduenas, L., 2010.Urban air pollution targets the dorsal vagal complex and dark chocolate offersneuroprotection. Int. J. Toxicol. 29, 604–615.

Vinson, J.A., Proch, J., Bose, P., Muchler, S., Taffera, P., Shuta, D., Samman, N.,Agbor, G.A., 2006. Chocolate is a powerful ex vivo and in vivo antioxidant,an anti-atherosclerotic agent in an animal model, and significant contribu-tor to antioxidants in European and American diets. J. Agric. Food Chem. 54,8071–8076.

Weinmann, S., Roll, S., Schwarzbachv, C., Vauth, C., Willich, S.N., 2010. Effects ofGinkgo biloba in dementia: systematic review and meta-analysis. BMC Geriatr.10, 14.

Whiting, D., 2001. Natural phenolic compounds 1900–2000: a bird’s eye view of acenturies chemistry. Nat. Prod. Rep. 18, 583–606.

Williams, R.J., Spencer, J.P., 2012. Flavonoids, cognition, and dementia: actions,mechanisms, and potential therapeutic utility for Alzheimer disease. Free Rad.Biol. Med. 52, 35–45.

Wilson, P.K., 2010. Centuries of seeking chocolate’s medicinal benefits. Lancet 376,158–159.

Wolfe, D., Shazzie, 2005. Naked Chocolate: The Astonishing Truth About the World’sGreatest Food. North Atlantic Books, Berkeley, CA.

Wollgast, J., Anklam, E., 2000. Review on polyphenols in Theobroma cacao: changesin composition during the manufacture of chocolate and methodology for iden-tification and quantification. Food Res. Int. 33, 423–447.

Wolz, M., Schleiffer, C., Klingelhöfer, L., Schneider, C., Proft, F., Schwanebeck, U.,Reichmann, H., Riederer, P., Storch, A., 2012. Comparison of chocolate to cacao-free white chocolate in Parkinson’s disease: a single-dose, investigator-blinded,placebo-controlled, crossover trial. J. Neurol. 259, 2447–2451.

Wolz, M., Kaminsky, A., Löhle, M., Koch, R., Storch, A., Reichmann, H., 2009.

Chocolate consumption is increased in Parkinson’s disease. Results from a self-questionnaire study. J. Neurol. 256, 488–492.

Yamada, T., Yamada, Y., Okano, Y., Terashima, T., Yokogoshi, H., 2009. Anxiolyticeffects of short- and long-term administration of cacao mass on rat elevatedT-maze test. J. Nutr. Biochem. 20, 948–955.