1 Mapping Brain Activity Induced by Olfaction of Virgin Olive Oil Aroma 2 Diego L. García-Gonz  alez,* ,† Jorge Vivancos, ‡ and Ram on Aparicio † 3 † Instituto de la Grasa (CSIC), Padre García Tejero 4, E-41012 Sevilla, Spain 4 ‡ Hospital San Juan de Dios, Avda. San Juan de Dios s/n, E-41930, Bormujos, Spain 5 ABSTRACT: The difficulty of explaining sensory descriptors of virgin olive oil aroma by the analysis of volatile compounds is 6 partially due to the subjective opinions of panelists and the lack of information of the neural mechanisms that ultimately produce a 7 sensory perception. In this study the technique of functional magnetic resonance imaging (fMRI) has been applied to study brain 8 activity during the smelling of virgin olive oil of different qualities. The volatile compounds of the samples were analyzed by solid- 9 phase microextraction gas chromatography to explain the differences in the aromas presented to the subjects during the fMRI 10 experiments. Comparing the pleasant and unpleasant aromas, the most evident differences in brain activity were found at the 11 anterior cingulate gyrus (Brodmann area 32) and at the temporal lobe (Brodmann area 38). The activations were also observed 12 when subjects smelled dilutions of heptanal and hexanoic acid, both compounds being responsible for off-flavors. Other areas were 13 inherent to the olfaction task (e.g., Brodmann area 10) and to the intensity of the aroma (Brodmann area 6). 14 KEYWORDS: virgin olive oil, volatiles, aroma, fMRI, brain imaging 15 16 17 ’ INTRODUCTION 18 Food aroma is regarded as one of the most important factors 19 determining consumer acceptability, and consequently, it has an 20 enormous economical impact on price. The economical impor- 21 tance of food aroma and its effect on overall quality explain the 22 increasing interest for new objective methods of aroma analysis. 23 Particularly, virgin olive oil aroma is also a sensory property that 24 must be characterized by law according to the official method of 25 sensory assessment. 1 This methodology, extensively described in 26 the International Olive Council (IOC) regulations, was devel- 27 oped and validated to overcome the subjective component of 28 panelist opinions by appropriate data processing. In addition to 29 the perfection in data management and the standardization of the 30 procedure, the sensory assessment by panelists still has the 31 drawback of including certain subjective information. The sub- 32 jective bias of sensory assessment is even more evident when the 33 scores given by panel tests from different countries are compared. 34 Consequently, intensive research has pursued the development 35 of methods for the analysis of the flavor compounds responsible 36 for the aroma 2,3 to acquire objective and univocal information of 37 sensory quality. However, the most updated methodologies of 38 flavor analysis still fail to reproduce the same information 39 provided by sensory assessment. Several reasons explain the 40 gap between instrumental analysis and sensory assessment. The 41 first reason is the high complexity of volatile compounds in virgin 42 olive oil and many other foods, which constitutes a resolution 43 challenge for the current chromatographic techniques. 4 The 44 second reason is the kinetic component of the flavor release 45 occurring during eating and swallowing and influenced by many 46 factors, such as saliva composition and mouth movements. These 47 factors modulate the final sensory perception, and they are not 48 taken into account in most of methodologies for flavor analysis. 49 Finally, little is known about the physiological mechanisms by 50 which flavor compounds result in neural activation and ulti- 51 mately give rise to sensory perception. The scrutiny of those 52 mechanisms by postreceptor studies with the main purpose of 53 explaining the sensory quality of foods has been scarcely ad- 54 dressed. The application of medical techniques such as positron 55 emission tomography (PET) and functional magnetic resonance 56 imaging (fMRI) in the service of food science has opened a new 57 research field in flavor chemistry that complements the chemi- 58 cal/sensory studies on food quality. 59 The acquisition of brain activity images with fMRI is based on 60 the magnetic resonance of protons in living tissues. 5 Neural 61 activity is associated with an increase in blood flow. As a cons- 62 equence, the concentration of deoxyhemoglobin decreases, 63 which is reflected by an increase in the relaxation time and the 64 magnetic resonance signal measured by fMRI. Data from fMRI 65 experiments are analyzed with a contrast analysis 6 that assumes 66 the hypothesis of greater activity during a cognitive process 67 compared to the rest state. 7 Thus, the stimulus is sequentially 68 presented to a subject alternating with rest periods in a block 69 design (or paradigm) in an ON/OFF frame. In addition to 70 assuming a lower brain activity during rest periods, another 71 important assumption is that the timing at which the neural 72 responses are registered matches the time speci fied in the 73 paradigm. 8 Furthermore, the stimulus needs to be presented to 74 the individual enough times to get a reliable statistical signifi- 75 cance. 9 Prior to studying the statistical differences between the 76 signals acquired at ON/OFF periods, the functional and struc- 77 tural images of the brain are processed to avoid artifacts due to 78 slight movements or other phenomena not related to the 79 cognitive process under study. 6 The image processing constitutes 80 the most time-consuming and tedious step in fMRI analysis, 81 which has led to intensive research oriented toward the devel- 82 opment of new software that is more powerful and versatile for all 83 the applications. 10 The software packages Analysis of Functional Received: May 26, 2011 Accepted: August 12, 2011 Revised: August 11, 2011 Journal of Agricultural and Food Chemistry | 3b2 | ver.9 | 22/8/011 | 12:1 | Msc: jf-2011-02106b | TEID: mec00 | BATID: 00000 | Pages: 10.45 ARTICLE pubs.acs.org/JAFC rXXXX American Chemical Society A dx.doi.org/10.1021/jf202106b | J. Agric. Food Chem. XXXX, XXX, 000–000