PERSPECTIVE ARTICLE

published: 20 March 2013

HUMAN NEUROSCIENCE doi: 10.3389/fnhum.2013.00087

Odor mental imagery in non-experts in odors: a paradox?

Jean-Pierre Royet 1*, Chantal Delon-Martin 2,3 and Jane Plailly 1
1
Olfaction: From Coding to Memory Team, Lyon Neuroscience Research Center, Centre National de la Recherche Scientifique, CNRS UMR 5292, INSERM U1028,
University Lyon 1, Lyon, France
2
INSERM, U836, Equipe Neuroimagerie Fonctionnelle et Perfusion Cérébrale, Grenoble, France
3
Institut des Neurosciences, Université Joseph Fourier, Grenoble, France

Edited by: In agreement with the theoretical framework stipulating that mental images arise from
Hans-Jochen Heinze, University of neural activity in early sensory cortices, the primary olfactory cortex [i.e., the piriform
Magdeburg, Germany
cortex (PC)] is activated when non-olfactory-experts try to generate odor mental images.
Reviewed by:
This finding strongly contrasts with the allegation that it is typically impossible to mentally
Hauke R. Heekeren, Freie
Universität Berlin, Germany imagine odors. However, other neurophysiological or cognitive processes engaged in the
Lluís Fuentemilla, University of endeavor of odor mental imagery such as sniffing, attention, expectation, and cross-modal
Barcelona, Spain interactions involve the PC and could explain this paradox. To unambiguously study the
*Correspondence: odor mental imagery, we first argued the need to investigate odor experts who have
Jean-Pierre Royet, Olfaction: From
learned to specifically reactivate olfactory percepts. We then assert the necessity to
Coding to Memory, Centre de
Recherche en Neurosciences de explore the network dedicated to this function by considering variations in both the activity
Lyon, CNRS UMR 5292, level and the connection strength of the areas belonging to this network as a function of
INSERM U1028, Université the level of expertise of the odor experts.
Lyon 1, 50 Avenue Tony Garnier,
69366 Lyon Cedex 07, France.
Keywords: odor mental imagery, piriform cortex, perfumer, vividness, functional reorganization
e-mail: [email protected]

IS THE INABILITY TO MENTALLY IMAGINE ODORS we observe functional brain modifications associated with these
A DOGMA? improvements?
Our ability to mentally imagine visual and auditory scenes or
motor actions has been widely demonstrated using behavioral POTENTIAL CAUSES OF PIRIFORM CORTEX ACTIVATION
and cerebral imaging studies (Jeannerod, 1995; Halpern and The PC can be activated by other processes than the mental
Zatorre, 1999; Kosslyn et al., 2001). However, with regard to olfac- imagery process itself, such as the neurophysiological or cogni-
tion, the widespread assertion is that it is very difficult for the tive processes engaged in the endeavor of odor mental imagery.
average person to mentally imagine odors. Several authors have First, recent studies have shown that, similar to odor perception,
even claimed that recalling physically absent odors is not possible olfactory imagery in naïve subjects is accompanied by increased
(Engen, 1982, 1991; Crowder and Schab, 1995; Herz, 2000), but respiratory amplitude (Bensafi et al., 2003; Kleemann et al., 2009)
some behavioral studies tend to take the opposite view (Elmes, and that, as a corollary, preventing sniffing during the mental
1998; Stevenson and Case, 2005). Two functional imagery stud- imagery of odor resulted in a poorer image vividness (Arshamian
ies are in favor of the existence of odor mental imagery because et al., 2008). Because sniffing is not only merely a stimulus carrier
the authors observed activation of the primary olfactory cortex but is also a part of the olfactory percept (Mainland and Sobel,
[i.e., the piriform cortex (PC)] in odor-untrained (naïve) subjects 2006), and because sniffing results in PC activation (Sobel et al.,
(Djordjevic et al., 2005; Bensafi et al., 2007). These results sup- 1998; Koritnik et al., 2009), the necessity of the involvement of
port the general view reported in studies on vision, audition, and such an olfactomotor system during olfactory imagery in naïve
motor processes: that similar neural networks are activated dur- subjects could explain the PC activation.
ing mental imagery and the actual perception of sensory stimuli Second, activation in the olfactory primary cortex can also be
(Kosslyn et al., 2001). influenced by top-down modulation factors, such as attention or
These findings are nevertheless surprising, as they show that expectation, as subjects are attentive to their olfactory environ-
activation of the PC during odor mental imagery is possible in ment while attempting to mentally imagine smell. In agreement
non-experts. Therefore, is the assertion of our inability to men- with the emerging view of selective attention in the primary
tally imagine odors a dogma? As the Middle Class Gentleman sensory processing of vision, audition, and somatosensation in
from J. B. Poquelin Molière unknowingly spoke prose, can we humans (Pugh et al., 1996; Gandhi et al., 1999; Carlsson et al.,
unknowingly mentally imagine odors? How can we reconcile the 2000; Kanwisher and Wojciulik, 2000; Petkov et al., 2004), Zelano
seemingly easy activation of the PC with the apparent difficulty et al. (2005) found that subjects were able to pay attention to their
to mentally imagine odors in naïve subjects? Is it possible that olfactory environment while ignoring their auditory environ-
the PC activation previously observed in non-experts (Djordjevic ment. Additionally, they found strong attentional modulation at
et al., 2005; Bensafi et al., 2007) is not associated with an effec- the earliest cortical level of olfactory processing, with the frontal
tive mental imagery process? Are we able to improve our capacity PC responding preferentially to attended sniffs over unattended
for olfactory mental imagination with training? If so, could sniffs. Recently, Zelano et al. (2011) reported that the instructions

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 1

Royet et al. Odor mental imagery and expertise

to prepare for an olfactory task are even sufficient to induce a suggest that such a process is active? In other words, what can
significant anticipatory response in this region. Rather than a be used to determine if vivid olfactory images come into our
general effect of attention, they highlighted that the activation mind? By evaluating the relationship between the psychophysi-
pattern dedicated to this expectation phase specifically reflects the cal evaluation of mental imagery abilities and brain activation,
attended odor. According to the authors, “the brain generates pre- Olivetti Belardinelli et al. (2009) found greater involvement of
dictive templates or ‘search images’ in posterior PC, with physical sensory-specific cortices in high- vs. low-vivid subjects for visual,
correspondence to odor-specific pattern representations, to augment gustatory, kinesthetic, tactile, and somatic imagery modalities but
olfactory perception.” not auditory or olfactory imagery. They concluded that the vivid-
Third, PC activation could result from cross-modal associa- ness scores related to olfactory imagery do not predict olfactory-
tive learning. A visual object previously paired with an unrelated specific activations, likely due to the difficulty in generating vivid
odor, when presented by itself, induces neural activity in the PC images of smells. This phenomenon could be explained by two
in the absence of odor stimulation (Gottfried et al., 2002, 2004). particularities of the olfactory system.
Although evoked after explicit associative learning, this reactiva- First, whereas visual, tactile, and auditory stimuli can be
tion mechanism is likely similar to the cross-modal activation decomposed into multiple components coded in feature maps,
phenomena observed in the following studies. Reading lips in such as color, line orientation, movement, or luminous inten-
the absence of any sound or simply reading words with auditory sity for visual stimuli, odors cannot be decomposed into multiple
meaning can activate auditory cortices (Calvert et al., 1997; Kiefer components. They are induced by chemical molecules, and even
et al., 2008). By the same token, food pictures activate gustatory the smallest modifications to these molecules can drastically
areas (Simmons et al., 2005), and reading words whose meaning change odor quality. For instance, cis- and trans-p-menth-8-ene
have strong olfactory associations automatically activates the PC stereoisomers have exactly the same molecular formula but are
(Gonzalez et al., 2006); the authors of these studies emphasized perceived as smelling similar to hydrocarbon or an orange,
that the activation resulted from conceptual reenactments and a respectively (Ohloff, 1971). A recent imaging study provided neu-
process of ignition of the semantic words, respectively, but not robiological evidence that we have a categorical (e.g., woody,
from a mental imagery process. minty), not structural (e.g., alcohol, ester), odor quality coding
Meyer and Damasio (Damasio and Meyer, 2008; Meyer in the posterior PC (Howard et al., 2009). These data support
and Damasio, 2009) proposed a model of cerebral functioning the view that our odor perception is more holistic than analytic,
designed to account for how representations are stored in mem- which does not allow for the progressive recall of odor mental
ory so that mental images can be re-experienced during recall. images through the gradual gathering of olfactory features and
They suggested that “retro-activation uses information available makes the generation of vivid odor images a difficult and even
in the association cortices and makes this information explicit impossible process.
by reconstructing maps in the early cortices.” They called these Second, although, as observed for other sensory modalities,
association cortices that enable the multiregional retro-activation olfactory knowledge can be acquired in naïve subjects through
of explicit maps in early sensorimotor cortices convergence- perceptual learning by simple prolonged exposure to odors
divergence zones (CDZs) and suggested that mirror and grand- (Li et al., 2006), the difficulty for non-experts to mentally imag-
mother neurons operate as CDZs. In olfaction, the orbitofrontal ine odors clearly differs from other sensory modalities, in which
cortex (OFC) and insula are association cortices that could be everyone is able to construct conscious vivid mental images and
potential CDZs. We recently showed that observing emotional can play the role of an expert. Therefore, the level of olfactory
facial expressions of disgust in others or feeling disgust oneself expertise has a strong influence on the ability to generate an odor
following odor inhalation activated the same region in the ante- mental image. Gilbert et al. (1998) found better scores in fra-
rior insula (Wicker et al., 2003), suggesting that a mirror-neuron grance experts than non-experts for the vividness of olfactory but
matching system operates for emotional expressions in this not visual images. Studying this specific population is the best
region. A cross-modal interaction was also observed at the neu- method to accurately identify the mental processes underlying the
ronal level in the monkey OFC with bimodal and even trimodal creation of olfactory images.
responses to taste, olfactory, and visual stimuli (Rolls and Baylis,
1994). Furthermore, CDZs can be areas of the motor circuitry, ODOR MENTAL IMAGERY AND EXPERTISE LEVEL
as brain regions involved in understanding others’ actions also Perfumers have learned to form olfactory sensory representa-
respond to olfactory cues (Rossi et al., 2008; Tubaldi et al., 2010). tions through daily practice and extensive training. They claim to
In summary, PC activation can be explained by top-down be able to produce odor perceptual images in the total absence
attentional processes or cross-modal sensorimotor interactions of odorants. Recently, we took advantage of the variability in
but not necessarily by an odor mental imagery process. This could expertise level between student and professional perfumers to
explain the paradoxical finding that the PC is activated in non- identify brain areas associated with olfactory mental imagery
experts who try to mentally imagine a smell, even though their (Plailly et al., 2012). Briefly, odor names were successively pre-
ability to mentally imagine odors is poor to non-existent. sented. For each name, the experts were asked whether they could
mentally imagine the odor as if they physically perceived it. We
VIVIDNESS OF OLFACTORY IMAGES observed clear differences in the testimonies between groups.
If PC activation is not a decisive criterion to evaluate whether sub- Student perfumers reported that odor mental imagery was highly
jects perform odor mental imagery, then what evidence would demanding, arduous, and fleeting, despite 2 years of training.

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 2

Royet et al. Odor mental imagery and expertise

A student reported as follows: “Yes, it requires being really con- associated with the activation decrease reveal that brain activa-
centrated, it’s difficult, really, to see it, to imagine it, to know tions are stronger when the difficulty to perform the task is higher
what it is, it comes as a flash [. . . ], it’s really hyper-fleeting. It (i.e., in perfumers at the beginning of their career). In the retrieval
takes time before coming, and when it does happen, it’s a frac- process taxonomy, this is termed “retrieval effort” and “refers to
tion of a second.” By contrast, professional smellers reported the level of processing resources deployed in the service of a retrieval
that they were able to rapidly evoke odors and maintain mental attempt” (Tulving, 1983).
images for 2–3 s. A renowned perfumer described his experience In addition to modulating activation levels, performance
as follows: “At the very same time as the word appears, the odor gains, and high image vividness can be related to changes
mentally comes out, at the same instant, it’s very fast, it’s of the in odor perceptual coding. First, because odor aversive learn-
second.” Another reported that “There are products that I actu- ing both enhances perceptual discrimination and updates odor
ally use all the time, the image is immediately coming, then this quality representations in the posterior PC (Li et al., 2008),
is crazy, because I have realized that raw materials that I use in we hypothesize that olfactory expertise is similarly associated
this moment, information comes straight away, this has stricken with a keen discrimination ability and heightened segrega-
me . . . ” This outstanding superiority of professionals over stu- tion of the odor-specific activity maps. Second, because the
dent experts, despite a full 2 years of training, highlights the slow retrieval from long-term memory involves the concerted activ-
development of expertise in perfumers (Schab and Cain, 1992) ity of distributed networks (Maguire et al., 2000; Frankland
and reinforces the idea that non-experts cannot mentally imagine and Bontempi, 2005; Moscovitch et al., 2006), the acquisition
odors. of olfactory experience could be associated with an increased
When professional perfumers generated an olfactory mental strength in the network connections dedicated to the expertise
image, we observed a signal decrease proportional to the length area. Accordingly, when perfumers created odor mental images,
of expertise in the posterior PC (Figure 1), hippocampus, OFC, the right middle frontal gyrus, a key region in the neural signa-
and middle frontal gyrus. The greater the level of expertise, the ture of retrieval (Lepage et al., 2000), was strongly coactivated
less these regions were activated. The idea of functional reorga- with olfactory and memory regions in professionals; however,
nization in response to expertise was proposed. We associated in students, this region was not or less coactivated (Figure 2).
this finding with performance gains and high image vividness Because the memory consolidation model called the “multiple
and emphasized the plasticity of the olfactory system occur- trace theory” indicates that the prefrontal cortex plays a cru-
ring in response to intensive training. The performance gains cial role in the posthippocampal recall of remote memories

FIGURE 1 | Functional data in odor experts. Significant negative structural scan of a professional perfumer. aPC, anterior piriform cortex;
correlations between the activation levels (recorded during mental imagery) pPC, posterior piriform cortex; y, coordinate along the antero-posterior axis
and length of expertise in the left and right aPC and right pPC represented of the brain; r, correlation coefficient; P, probability value. Adapted from
in red on coronal sections of a normalized, T1-weighted, unsmoothed, Plailly et al. (2012).

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 3

Royet et al. Odor mental imagery and expertise

FIGURE 2 | Coactivation in the odor mental imagery network. Correlation between the middle frontal gyrus and the rest of the network (red arrows)
matrixes depicting functional coactivation between signal time courses of 22 was higher in professional than student experts (Wilcoxon signed ranks test:
pairs of regions of interest (right column) for olfactory mental imagery events Z = 3.055, p = 0.0022). In contrast, coactivations between the precuneus
in (A) student and professional experts (Plailly et al., 2012). Each cell indicates and other areas (blue arrow) was lower in professional than student experts
the group’s mean correlation coefficient that was computed between the (Z = 2.619, p = 0.0088). Abbreviations: MFG, Medial frontal gyrus; SFG,
activation signal of a pair of ROIs. Mean correlation values are shown at the Superior frontopolar gyrus; OFG, Orbitofrontal gyrus; aPC, anterior piriform
range of −1 (dark blue) to 1 (dark red). The cells depicted in the diagonal of cortex; pPC, posterior piriform cortex; Amy, amygdala; aHippo, anterior
each matrix represent the correlations between the activation level of each hippocampus; ParaH, parahippocampal gyrus; ITG, Inferior temporal gyrus;
region and itself. (B) The student coactivation matrix was subtracted from the MOG, Middle occipital gyrus; PostC, posterior central gyrus; Precu,
professional coactivation matrix. This operation shows that the coactivations precuneus; Cere, cerebellum.

(Frankland and Bontempi, 2005; Takashima et al., 2006), we pro- brain network dedicated to this function, depending on the sub-
pose that the middle frontal gyrus could fill this role during the jects’ expertise. By the same token, it would be interesting to
recall of an olfactory percept by ensuring an optimal top-down study the dynamic of the network involved in generating mental
reactivation of the PC. Another major difference in our study images of odor. For instance, methods exploring effective con-
between both groups was the strong coactivation between the nectivity, such as dynamic causal modeling (Friston et al., 2003),
precuneus and other regions involved in mental imagery in stu- could be used to test whether the relationship between higher
dents, but not in professionals (Figure 2). This disparity makes cortical regions and the PC is influenced by the vividness of
sense because the precuneus is an area that pertains to the supe- the olfactory image generated. Furthermore, other topics for
rior parietal lobe, which is active when the need for top-down future research directions are also conceivable. Howard et al.
assistance to memory retrieval is maximal (Ciaramelli et al., (2009) have shown that the coding of odor categorical percep-
2008). tion is regionally specific for the posterior PC. They used the
multivariate techniques that are based on the pattern of voxels
CONCLUSIONS AND FUTURE DIRECTIONS activated for a specific stimulus in a specific participant, which
Except for the small minority of individuals whose work leads allows the characterization of how (rather than just where) the
them to train their olfactory abilities (such as perfumers, chefs, perceptual information is represented in the brain. Using the
flavorists, and oenologists), most individuals claim not to be able same multivariate techniques, a challenge for future research
to create an odor mental image, and thus to have the feeling would be to investigate the coding of mental images of odor
of perceiving a smell in the nose. However, recent experiments in olfactory experts and answer the following question: is the
take the opposite view and support the existence of odor men- pattern of activation dedicated to the perception of an odor
tal imagery ability in non-experts, using the observation of PC similar to the one dedicated to the imagination of the same
activation during this process as an indicator. We suggest that odor? In a similar vein, Howard et al. (2009) have demon-
this paradoxical finding could have several explanations. The PC strated that the anterior and posterior PC and OFC contain
may be incidentally reactivated during sniffing, odor expecta- ensemble representations of individual odorants (odorant iden-
tion or attention, cross-modal recall of information previously tity). However, learning influences odor perceptual coding in
linked with odor through associative learning, or the ignition the posterior PC only (Li et al., 2008). These same multivari-
of semantic words. Thus, the observation of PC activation in ate pattern-based techniques could be used to compare activation
subjects attempting to generate odor mental images does not patterns between odor experts and naïve subjects to test whether
irrefutably indicate that the odor percept has been reactivated odor percepts are coded with more complex patterns in experts
and that the odor has been mentally imagined. Other indices than in naïve subjects and if additional regions, such as the OFC,
must be taken into account, such as the self-reported ability are involved. Lastly, several studies in humans indicate structural
of subjects to imagine odors, the variations of both the level modifications in the brains of musicians and athletes as a con-
of activity in olfactory areas and connection strength in the sequence of learning and training (Jancke, 2009). Whether there

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 4

Royet et al. Odor mental imagery and expertise

are also structural changes in odor experts remains to be ACKNOWLEDGMENTS

determined. The voxel-based morphometry technique could be The authors are grateful to the Program “Investissements
used to measure variations in gray and white matter volume in d’Avenir”—launched by the French Government and
the brain, and the diffusion tensor imaging technique could allow operated by the National Research Agency (ANR)—for
for the exploration of the architecture of white matter and axonal financial support to the LabEx CORTEX of Université
connectivity. de Lyon.

REFERENCES and remote memories. Nat. Rev. Kiefer, M., Sim, E. J., Herrnberger, semantic and spatial mem-
Arshamian, A., Olofsson, J. K., Jönsson, Neurosci. 6, 119–130. B., Grothe, J., and Hoenig, ory. Curr. Opin. Neurobiol. 16,
F. U., and Larsson, M. (2008). Sniff Friston, K. J., Harrison, L., and Penny, K. (2008). The sound of con- 179–190.
your way to clarity: the case of W. (2003). Dynamic causal mod- cepts: four markers for a link Ohloff, G. (1971). L’odorat et la
olfactory imagery. Chem. Percept. 1, elling. Neuroimage 19, 1273–1302. between auditory and conceptual forme des molécules. Recherche 18,
242–246. Gandhi, S. P., Heeger, D. J., and brain systems. J. Neurosci. 28, 1068–1970.
Bensafi, M., Porter, J., Pouliot, S., Boynton, G. M. (1999). Spatial 12224–12230. Olivetti Belardinelli, M., Palmiero,
Mainland, J., Johnson, B., Zelano, attention affects brain activity Kleemann, A. M., Kopietz, R., Albrecht, M., Sestieri, C., Nardo, D., Di
C., et al. (2003). Olfactomotor activ- in human primary visual cortex. J., Schopf, V., Pollatos, O., Schreder, Matteo, R., Londei, A., et al. (2009).
ity during imagery mimics that dur- Proc. Natl. Acad. Sci. U.S.A. 96, T., et al. (2009). Investigation An fMRI investigation on image
ing perception. Nat. Neurosci. 6, 3314–3319. of breathing parameters during generation in different sensory
1142–1144. Gilbert, A. N., Crouch, M., and Kemp, odor perception and olfac- modalities: the influence of vivid-
Bensafi, M., Sobel, N., and Khan, R. M. S. E. (1998). Olfactory and visual tory imagery. Chem. Senses 34, ness. Acta Psychol. (Amst.) 132,
(2007). Hedonic-specific activity mental imagery. J. Ment. Imag. 22, 1–9. 190–200.
in piriform cortex during odor 137–146. Koritnik, B., Azam, S., Andrew, C. Petkov, C. I., Kang, X., Alho, K.,
imagery mimics that during odor Gonzalez, J., Barros-Loscertales, A., M., Leigh, P. N., and Williams, Bertrand, O., Yund, E. W., and
perception. J. Neurophysiol. 98, Pulvermuller, F., Meseguer, V., S. C. (2009). Imaging the brain Woods, D. L. (2004). Attentional
3254–3262. Sanjuan, A., Belloch, V., et al. during sniffing: a pilot fMRI modulation of human auditory cor-
Calvert, G. A., Bullmore, E. T., (2006). Reading cinnamon activates study. Pulm. Pharmacol. Ther. 22, tex. Nat. Neurosci. 7, 658–663.
Brammer, M. J., Campbell, R., olfactory brain regions. Neuroimage 97–101. Plailly, J., Delon-Martin, C., and Royet,
Williams, S. C., McGuire, P. K., 32, 906–912. Kosslyn, S. M., Ganis, G., and J. P. (2012). Experience induces
et al. (1997). Activation of auditory Gottfried, J. A., O’Doherty, J., and Thompson, W. L. (2001). Neural functional reorganization in brain
cortex during silent lipreading. Dolan, R. J. (2002). Appetitive foundations of imagery. Nat. Rev. regions involved in odor imagery in
Science 276, 593–596. and aversive olfactory learn- Neurosci. 2, 635–642. perfumers. Hum. Brain Mapp. 33,
Carlsson, K., Petrovic, P., Skare, S., ing in humans studied using Lepage, M., Ghaffar, O., Nyberg, L., and 224–234.
Petersson, K. M., and Ingvar, M. event-related functional magnetic Tulving, E. (2000). Prefrontal cor- Pugh, K. R., Offywitz, B. A., Shaywitz,
(2000). Tickling expectations: neu- resonance imaging. J. Neurosci. 22, tex and episodic memory retrieval S. E., Fulbright, R. K., Byrd,
ral processing in anticipation of a 10829–10837. mode. Proc. Natl. Acad. Sci. U.S.A. D., Skudlarski, P., et al. (1996).
sensory stimulus. J. Cogn. Neurosci. Gottfried, J. A., Smith, A. P., Rugg, 97, 506–511. Auditory selective attention: an
12, 691–703. M. D., and Dolan, R. J. (2004). Li, W., Howard, J. D., Parrish, T. fMRI investigation. Neuroimage 4,
Ciaramelli, E., Grady, C. L., and Remembrance of odors past: B., and Gottfried, J. A. (2008). 159–173.
Moscovitch, M. (2008). Top-down human olfactory cortex in cross- Aversive learning enhances percep- Rolls, E. T., and Baylis, L. L. (1994).
and bottom-up attention to mem- modal recognition memory. Neuron tual and cortical discrimination of Gustatory, olfactory, and visual
ory: a hypothesis (AtoM) on the role 42, 687–695. indiscriminable odor cues. Science convergence within the primate
of the posterior parietal cortex in Halpern, A. R., and Zatorre, R. J. 319, 1842–1845. orbitofrontal cortex. J. Neurosci. 14,
memory retrieval. Neuropsychologia (1999). When that tune runs Li, W., Luxenberg, E., Parrish, T., and 5437–5452.
46, 1828–1851. through your head: a PET inves- Gottfried, J. A. (2006). Learning Rossi, S., De Capua, A., Pasqualetti, P.,
Crowder, R. G., and Schab, F. R. (1995). tigation of auditory imagery for to smell the roses: experience- Ulivelli, M., Fadiga, L., Falzarano,
“Imagery for odors,” in Memory for familiar melodies. Cereb. Cortex dependent neural plasticity in V., et al. (2008). Distinct olfactory
Odors, eds R. G. Crowder and F. 9, 697–704. human piriform and orbitofrontal cross-modal effects on the human
R. Schab (Hillsdale, NJ: Erlbaum), Herz, R. S. (2000). Verbal coding in cortices. Neuron 52, 1097–1108. motor system. PLoS ONE 3:e1702.
93–107. olfactory versus nonolfactory cogni- Maguire, E. A., Mummery, C. J., and doi: 10.1371/journal.pone.0001702
Damasio, A., and Meyer, K. (2008). tion. Mem. Cognit. 28, 957–964. Buchel, C. (2000). Patterns of Schab, F. R., and Cain, W. S. (1992).
Behind the looking-glass. Nature Howard, J. D., Plailly, J., Grueschow, hippocampal-cortical interaction “Memory for odors,” in The Human
454, 167–168. M., Haynes, J. D., and Gottfried, J. dissociate temporal lobe mem- Sense of Smell, eds D. G. Laing, R.
Djordjevic, J., Zatorre, R. J., Petrides, A. (2009). Odor quality coding and ory subsystems. Hippocampus 10, L. Doty, and W. Breipohl (Berlin:
M., Boyle, J. A., and Jones-Gotman, categorization in human posterior 475–482. Springer-Verlag), 217–240.
M. (2005). Functional neuroimag- piriform cortex. Nat. Neurosci. 12, Mainland, J., and Sobel, N. (2006). The Simmons, W. K., Martin, A., and
ing of odor imagery. Neuroimage 932–938. sniff is part of the olfactory percept. Barsalou, L. W. (2005). Pictures of
24, 791–801. Jancke, L. (2009). The plastic human Chem. Senses 31, 181–196. appetizing foods activate gustatory
Elmes, D. G. (1998). Is there an inner brain. Restor. Neurol. Neurosci. 27, Meyer, K., and Damasio, A. (2009). cortices for taste and reward. Cereb.
nose? Chem. Senses 23, 443–445. 521–538. Convergence and divergence in a Cortex 15, 1602–1608.
Engen, T. (1982). The Perception of Jeannerod, M. (1995). Mental neural architecture for recognition Sobel, N., Prabhakaran, V., Desmond,
Odors. New York, NY: Academic imagery in the motor context. and memory. Trends Neurosci. 32, J. E., Glover, G. H., Goode, R.
Press. Neuropsychologia 33, 1419–1432. 376–382. L., Sullivan, E. V., et al. (1998).
Engen, T. (1991). Odor Sensation and Kanwisher, N., and Wojciulik, E. Moscovitch, M., Nadel, L., Winocur, Sniffing and smelling: sepa-
Memory. New York, NY: Praeger. (2000). Visual attention: insights G., Gilboa, A., and Rosenbaum, rate subsystems in the human
Frankland, P. W., and Bontempi, B. from brain imaging. Nat. Rev. R. S. (2006). The cognitive neu- olfactory cortex. Nature 392,
(2005). The organization of recent Neurosci. 1, 91–100. roscience of remote episodic, 282–286.

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 5

Royet et al. Odor mental imagery and expertise

Stevenson, R. J., and Case, T. I. odors, understanding actions. Soc. Zelano, C., Mohanty, A., and Gottfried, Citation: Royet J-P, Delon-Martin C and
(2005). Olfactory imagery: a Neurosci. 6, 1–17. J. A. (2011). Olfactory predictive Plailly J (2013) Odor mental imagery in
review. Psychon. Bull. Rev. 12, Tulving, E. (1983). Elements of Episodic codes and stimulus templates in pir- non-experts in odors: a paradox? Front.
244–264. Memory. Oxford: Clarendon. iform cortex. Neuron 72, 178–187. Hum. Neurosci. 7:87. doi: 10.3389/
Takashima, A., Petersson, K. M., Wicker, B., Keysers, C., Plailly, J., Royet, fnhum.2013.00087
Rutters, F., Tendolkar, I., Jensen, J. P., Gallese, V., and Rizzolatti, G. Conflict of Interest Statement: The Copyright © 2013 Royet, Delon-Martin
O., Zwarts, M. J., et al. (2006). (2003). Both of us disgusted in My authors declare that the research and Plailly. This is an open-access
Declarative memory consolidation insula: the common neural basis of was conducted in the absence of any article distributed under the terms of the
in humans: a prospective functional seeing and feeling disgust. Neuron commercial or financial relationships Creative Commons Attribution License,
magnetic resonance imaging study. 40, 655–664. that could be construed as a potential which permits use, distribution and
Proc. Natl. Acad. Sci. U.S.A. 103, Zelano, C., Bensafi, M., Porter, J., conflict of interest. reproduction in other forums, provided
756–761. Mainland, J., Johnson, B., Bremner, the original authors and source are
Tubaldi, F., Turella, L., Pierno, A. E., et al. (2005). Attentional modu- Received: 09 July 2012; accepted: 02 credited and subject to any copyright
C., Grodd, W., Tirindelli, R., and lation in human primary olfactory March 2013; published online: 20 March notices concerning any third-party
Castiello, U. (2010). Smelling cortex. Nat. Neurosci. 8, 114–120. 2013. graphics etc.

Frontiers in Human Neuroscience www.frontiersin.org March 2013 | Volume 7 | Article 87 | 6