CHAPTER
Olfactory Insights into
Sleep-Dependent Learning
and Memory
12
Laura K. Shanahan, Jay A. Gottfried1
Department of Neurology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA
1
Corresponding author: Tel.: þ1-312-503-3185; Fax: þ1-312-503-0872,
e-mail address: [email protected]
Abstract
Sleep is pervasive throughout most of the animal kingdom—even jellyfish and honeybees do it.
Although the precise function of sleep remains elusive, research increasingly suggests that
sleep plays a key role in memory consolidation. Newly formed memories are highly labile
and susceptible to interference, and the sleep period offers an optimal window in which memories can be strengthened or modified. Interestingly, a small but growing research area has begun
to explore the ability of odors to modulate memories during sleep. The unique anatomical organization of the olfactory system, including its intimate overlap with limbic systems mediating
emotion and memory, and the lack of a requisite thalamic intermediary between the nasal
periphery and olfactory cortex, suggests that odors may have privileged access to the brain during sleep. Indeed, it has become clear that the long-held assumption that odors have no impact on
the sleeping brain is no longer tenable. Here, we summarize recent studies in both animal and
human models showing that odor stimuli experienced in the waking state modulate olfactory
cortical responses in sleep-like states, that delivery of odor contextual cues during sleep can
enhance declarative memory and extinguish fear memory, and that olfactory associative learning
can even be achieved entirely within sleep. Data reviewed here spotlight the emergence of a new
research area that should hold far-reaching implications for future neuroscientific investigations
of sleep, learning and memory, and olfactory system function.
Keywords
sleep, olfaction, olfactory system, human brain, piriform cortex, olfactory bulb, associative
learning, memory reactivation, emotion, cognition
1 INTRODUCTION
Behavioral states resembling sleep occur in the great majority of vertebrate and invertebrate species. Vertebrate sleep may have been present in the earliest jawed
fishes 435 million years (My) ago, and based on fossil records, invertebrate sleep
Progress in Brain Research, Volume 208, ISSN 0079-6123, http://dx.doi.org/10.1016/B978-0-444-63350-7.00012-7
© 2014 Elsevier B.V. All rights reserved.
309
310
CHAPTER 12 Smell, Sleep, and Memory
may date back to ancestors of the chambered nautilus 543 My ago (Kavanau, 2006). A
brain is highly useful for going to sleep, but it is not necessary. For example, the
highly lethal box jellyfish, Chironex fleckeri, which likely emerged a few million
years ago, lacks a central nervous system but still takes the time to hover motionless
over the Australian sea floor for up to 15 h a day (Kavanau, 2006; Seymour et al.,
2004). The implication is that throughout evolution there has been a strong selective
pressure to maintain sleep in an animal’s behavioral repertoire. Even animals that
appear to have escaped the need for sleep, such as whales and dolphins that
swim continuously, demonstrate unihemispheric sleep (Lyamin et al., 2002;
Mukhametov et al., 1977). The next section of this chapter will consider some of
the proposed behavioral, physiological, and cognitive benefits that sleep might provide.
We humans are utterly familiar with sleep, committing one-third of each day to
the act, but we remain relatively ignorant about why we sleep. Throughout history,
philosophers, artists, poets, and priests have attempted to make sense of sleep, long
before psychoanalysis and neuroscience came into vogue. In ancient religion and
mythology, spiritual embodiments of sleep were commonplace, and the gods and
goddesses of sleep took many names: Ah Uaynih (Mayan), Caer Ibormeith (Irish/
Celtic), Hypnos and his son Morpheus (Greek), Jum Sum (Chinese), Njörun (Norse),
Somnus (Roman), Tutu (Egyptian), Urmya (Vedic/Hindu), and Zakar (Babylonian).
The topic of sleep featured prominently in the Song of Songs (Song of Solomon) in
the Old Testament, with passages describing a bride’s tormented dreams of her lover
(see parts 3 and 5). During the fifteenth and sixtieth centuries, Renaissance painters
portrayed the meaning of sleep in highly imaginative ways, providing religious,
moral, mythical, and psychological depictions of sleep and dreaming: the divine
and the damned, visions and transcendence, night and death, and the faintly erotic
and the more erotic (Cecchi et al., 2013). Perhaps anticipating the future of olfactory
sleep research (and the main theme of this chapter), the Venetian Renaissance painter
Lorenzo Lotto (ca. 1480–1556/1557) depicted a young maiden napping against a
tree, while a winged putto scatters aromatic (?) white flowers in a flowing stream
down onto her head (Fig. 1), no doubt an early characterization of the potential
for odor stimuli to influence the contents of sleep—in this case, the dream-like evocation of a pair of grotesque and debauched satyrs. A main goal of this chapter will be
to present recent animal and human findings that extend Lotto’s incipient idea that
odors have a direct influence on sleep physiology, memory, and behavior.
This chapter continues with a brief review of the phenomenology, physiology,
and proposed functions of sleep, followed by an in-depth discussion of the interface
between smells and sleep, with specific focus on both human and animal studies, and
the concept of using olfactory-targeted memory reactivation (TMR) to influence
memory consolidation in the sleeping brain. Because there is not enough space here
to provide a detailed overview of the neuroscience of sleep or odor processing in
the mammalian olfactory system, the reader may wish to consult many recent
reviews on these topics (sleep: Brown et al., 2012; Fuller et al., 2006; Rasch and
Born, 2013; Walker and Stickgold, 2006; and olfaction: Arzi and Sobel, 2011;
Gottfried, 2010; Murthy, 2011; Wachowiak, 2011; Wilson and Sullivan, 2011).
1 Introduction
FIGURE 1
Lorenzo Lotto (ca. 1480–1556/1557), Allegory of Chastity, oil on canvas, ca. 1505, Samuel H.
Kress Collection, National Gallery of Art, Washington, DC.
Downloaded at http://commons.wikimedia.org/wiki/File:Lorenzo_Lotto_077.jpg on January 10, 2014.
311
312
CHAPTER 12 Smell, Sleep, and Memory
2 SLEEP: A BRIEF OVERVIEW
Sleep can be loosely defined as a condition of reversible unconsciousness accompanied by relative inactivity and unresponsiveness. (This can be contrasted to pathological states of reversible unconsciousness that are accompanied by excessive
activity, as in the case of sleepwalking or temporal lobe seizures.) It seems counterintuitive that such a behaviorally vulnerable state could be advantageous, but the
preservation of sleep over the course of evolution, as well as its prevalence throughout the animal kingdom, suggests that sleep must have its uses (Cirelli and Tononi,
2008). Additional lines of evidence indicate that sleep is not only advantageous but
also critical to survival. For instance, sleep is homeostatically regulated; that is, sleep
loss results in longer and/or deeper sleep during the subsequent sleep period. After
extended periods of sleep deprivation, sleep begins to intrude on wakefulness, and
spectral analysis of electroencephalogram (EEG) recordings in rats reveals periods
in which low-frequency oscillations characteristic of sleep invade bouts of wake
(Franken et al., 1991; Friedman et al., 1979). Prolonged sleep deprivation also impairs
cognitive function (Thomas et al., 2000; Van Dongen et al., 2003), and, in extreme
cases, can be fatal (Rechtschaffen and Bergmann, 2002; Stephenson et al., 2007).
2.1 Sleep Phenomenology and Architecture
Sleep can be parsed into distinct electrophysiological stages, and sleep architecture
refers to the pattern of those stages over the course of the night (Fig. 2). Originally
classified by Rechtschaffen and Kales, each stage can be identified by its unique
neuronal signature as recorded by EEG (Rechtschaffen and Kales, 1968). The characteristics of neuronal activity accompanying each stage are summarized below (Iber
et al., 2007; Inostroza and Born, 2013).
During the wake state with eyes closed, brain activity is dominated by a posterior
alpha rhythm (8.5–13 Hz). As drowsiness ensues, the alpha rhythm becomes increasingly attenuated, marking the onset of stage 1 sleep, which can include vertex sharp
waves. As the subject slips further into stage 2 sleep, the EEG is typified by two features: biphasic negative/positive deflections (“K-complexes”) that stand out from
background activity; and brief trains of 12–15 Hz oscillations (“spindles”) generated
by thalamocortical loops. The duration of K-complexes and spindles must exceed
0.5 s to be classified as such. Stage 3 sleep, commonly referred to as slow-wave sleep
(SWS), consists of low-frequency (0.5–4 Hz), high-amplitude (>75 mV) delta oscillations, upon which spindles are often superimposed (Fig. 2B, left). Recent research
also shows that stage 3 sleep contains sharp-wave “ripples,” transient high-frequency
bursts of activity (100–250 Hz). Ripples have been documented chiefly in the hippocampus and may play a key role in memory consolidation (Ego-Stengel and
Wilson, 2010; Girardeau et al., 2009), but are extremely difficult to identify using
surface EEG techniques. Rapid eye movement sleep (REM) consists of highfrequency, low-amplitude waves similar to those observed in the wake state
(Fig. 2B, right), and may include 4–6 Hz trains of sharply-angled, serrated waves
2 Sleep: A Brief Overview
FIGURE 2
Sleep physiology. (A) Human sleep architecture over the course of typical nighttime sleep,
depicting wake, REM, and nREM stages 1–3. Note that stage 3, or SWS, is more frequent in
the first half of the night. (B) Surface EEG recordings in humans (top row) demonstrate the
low-frequency, high-amplitude delta oscillations characteristic of SWS punctuated by
spindles (12–15 Hz). REM is characterized by high-frequency, low-amplitude activity. Local
field potential recordings from rat hippocampus (bottom row) demonstrate sharp-wave
ripples during nREM consisting of rapid depolarizations in CA3 overlaid by ultra-highfrequency ripple activity (100–250 Hz) generated in CA1. In rodents, theta activity (4–8 Hz) is
prominent during REM.
Adapted and modified from Inostroza and Born (2013).
(“sawtooth waves”). As per its name, REM is associated with increased oculomotor
activity on electrooculogram recordings, as well as muscle atonia interspersed with
transient bursts of muscle activity on electromyogram (EMG) recordings.
Over the course of overnight sleep, these sleep stages progress in a cyclic manner,
with each cycle lasting around 90 min. Humans complete four to five cycles each
night (Fig. 2A). Of note, cycles in the first half of the night are predominated by
SWS, whereas cycles in the second half consist chiefly of REM. One important
implication for asymmetrical sleep architecture is that short naps contain mostly
SWS, which holds potential relevance for sleep-based experimental designs in which
313
314
CHAPTER 12 Smell, Sleep, and Memory
SWS is the main focus. Stages 1, 2, and 3 are cumulatively referred to as non-REM
sleep (nREM), and in rodents, there is no distinction between individual nREM
stages. In the context of rodent studies, SWS and nREM are synonymous.
2.2 The Proposed Functions of Sleep
Sleep likely serves many restorative and homeostatic functions. In the simplest formulation, physical activity and even mental effort are energy-demanding processes, and
what better way to replenish one’s energy stores than to take a nap—though research
intriguingly suggests that energy consumption is similar during wake and sleep states
(Dworak et al., 2010). Sleep can exert restorative effects at the metabolic (Knutson
et al., 2007; Morselli et al., 2012; Van Cauter et al., 2008) and immunological levels
(Besedovsky et al., 2012; Lange et al., 2010). The observation that animals sleep more
in the postnatal and juvenile periods than as adults has given rise to the interesting possibility that sleep is especially critical for brain development (Wang et al., 2011).
One influential hypothesis is that sleep is essential for synaptic homeostasis
(Tononi and Cirelli, 2003, 2006). The basic idea is that over the course of a typical
day, an animal experiences many new episodes and events (some meaningful, others
irrelevant), all of which will induce some degree of synaptic plasticity, with molecular, physiological, and anatomical modifications in the brain. Sleep affords a window in which synaptic downscaling can take place, effectively cleaning up all of
the synaptic noise that accrued in the waking state. Put differently, the physiological
excesses of learning and plasticity that inevitably occur in the waking state are mitigated through mechanisms of synaptic homeostasis employed during sleep. This
process would have important anatomical and functional consequences, ensuring
efficient use of the finite space within the brain, and preserving only the most robust
synaptic changes for subsequent consolidation.
Another important idea is that the function of sleep is to promote memory consolidation. In fact, this concept is not incompatible with the synaptic homeostasis
hypothesis, insofar as pruning and downscaling of irrelevant synapses should help
stabilize relevant memory traces for subsequent consolidation. The question of
whether sleep can enhance memory has inspired countless research studies, and despite strong supporting data, the underlying mechanisms remain widely contested.
In a landmark experiment, Jenkins and Dallenbach (1924) conducted a memory
study that required subjects to memorize nonsense syllables and recall them after
an interval of sleep or wake (Jenkins and Dallenbach, 1924). Notably, the sleep group
demonstrated superior recall. The investigators speculated that sleep augments memory function simply by shielding memories from interference, a theory that garnered
support for many years and continues to hold sway in the literature. In this scenario,
sleep is viewed as a passive agent, enabling memory consolidation to take place without itself playing an active role.
However, recent behavioral and neurobiological studies provide compelling
evidence that sleep may play a more active role in memory consolidation
(Ellenbogen et al., 2006). Early support for this idea came from rodent studies, which
3 The Interaction Between Smells and Sleep: Human Studies
demonstrated that the same hippocampal place cells activated during a spatial exploration task were also activated during subsequent sleep (Pavlides and Winson, 1989), and
particularly during SWS (Wilson and McNaughton, 1994). These findings were compatible with the idea that reactivation or replay of information processing during sleep
is critical for memory consolidation. Since these seminal findings, and with many further
contributions from both animal and human research (Deuker et al., 2013; Tambini and
Davachi, 2013; Tambini et al., 2010), a working model of sleep-dependent memory consolidation has begun to emerge: memory traces are initially encoded in the hippocampus,
where they are highly labile; during sleep, these memory traces are repeatedly reactivated, leading to stabilization and redistribution to extra-hippocampal cortical networks
for long-term storage. It is thought that the slow oscillations that occur during SWS
underlie reactivation of memories, whereas thalamocortical spindles and sharp-wave
ripples may be involved in their redistribution (Rasch and Born, 2013). The recent implementation of TMR in human and animal studies (Antony et al., 2012; Bendor and
Wilson, 2012; Diekelmann et al., 2011, 2012; Fuentemilla et al., 2013; Hauner et al.,
2013; Oudiette et al., 2013a; Rasch et al., 2007; Rolls et al., 2013; Rudoy et al.,
2009; van Dongen et al., 2012) has offered an exciting new method to test the active
memory consolidation hypothesis and will be discussed in detail in Section 5.
3 THE INTERACTION BETWEEN SMELLS AND SLEEP: HUMAN
STUDIES
For thousands of years, humans have used odors as fragrant sleeping draughts. In
ancient Egypt, kyphi, an odorous resin of calamus, henna, spikenard, frankincense,
myrrh, cinnamon, cypress, and terebinth (pistachio resin), was burned at the altar in
Heliopolis to induce sleep and enhance dreams (Keville and Green, 1995). It seems
quite probable that the smoke arising from many of these smoldering plants and herbs
contained sedative-hypnotic or even hallucinogenic properties: scholars have suggested that at the Oracle of Delphi in ancient Greece, the fantastic visions and prophecies divined by the priestesses were due to the high concentrations of plant-derived
fumes wafting into the inner sanctum (Pennacchio et al., 2010).
Empirical evidence for the ability of smells to promote sleep has been slow to
emerge and modest in content. This research has mostly focused on the potential soporific role of lavender odor, likely motivated by long-standing anecdotal reports
(including one from Queen Victoria herself!). Studies suggest that exposure to lavender before going to the sleep can enhance sleep quality in patients with mild insomnia (Lewith et al., 2005), extend the duration of the first cycle of SWS (Goel
et al., 2005), improve vigor the following morning (Goel et al., 2005), and induce
deeper sleep for napping infants (Field et al., 2008). The Goel group conducted a
follow-up study in which they delivered peppermint odor prior to nighttime sleep,
the prediction being that peppermint odor (which is considered to be stimulating)
would disrupt sleep (Goel and Lao, 2006). Instead, exposure to peppermint odor increased the duration of SWS for those subjects who perceived the odor to be intense.
315
316
CHAPTER 12 Smell, Sleep, and Memory
Further research is clearly necessary to establish the actual mechanisms by which
certain odors influence sleep and arousal, but these data provide a scientific foundation for the use of “aromatherapy” in improving sleep.
Limited research has also examined the impact of sleep on odor perception.
Killgore and McBride (2006) showed that after 24 h of sleep deprivation, odor identification on the University of Pennsylvania Smell Identification Test (Doty et al.,
1984) was impaired. Intact performance on a test of sustained attention and executive
function suggested that task difficulty or decreased vigilance per se could not explain
the olfactory perceptual deficits. However, these results remain inconclusive, in that
those participants reporting greater subjective sleepiness paradoxically performed
better on the odor identification test. The group conducted a follow-up study confirming that 54-h sleep deprivation hindered odor identification to the same extent
as 24-h sleep deprivation (McBride et al., 2006). Though far from definitive, the
above studies illustrate the reciprocal influences between odors and sleep, and intimate that odors delivered during sleep may have a profound impact on the sleeper.
3.1 Response to Odors During Sleep
The experience of awakening to the sound of a quiet voice or a gentle tap on the shoulder is a common one, but it is unusual to be aroused by even the most pungent of odors.
That said, the middle of the night is not usually punctuated by the sudden presence of
an odor (with a few flatulent exceptions), so the question of whether olfactory stimuli
can be detected during sleep has not really been put to the test. Indeed, compared to
other sensory modalities, research investigating olfactory-evoked arousal has been
scant (Velluti, 1997). However, based on the distinctive anatomical organization of
the olfactory system—in which odor information processing in olfactory cortex does
not need to be routed through the thalamus (Carmichael et al., 1994; Ray and Price,
1992; Russchen et al., 1987; Tanabe et al., 1975; Yarita et al., 1980)—one could plausibly expect that the sleeping brain should be responsive to odors.
A study by Badia and colleagues was the first to characterize olfactory sensitivity
in humans during sleep (Badia et al., 1990). Either peppermint odor or clean air (control) was delivered in 3-min blocks over the course of stage 2 sleep, while odorinduced behavioral arousal (indexed via button-press), EEG and EMG activity, heart
rate, and respiratory rate were concurrently recorded. Odor delivery (vs. control) was
associated with a marginal increase in behavioral response, along with increased
heart rate, decreased EMG activity, and increased EEG “speeding” (high-frequency
EEG bursts lasting less than 10 s). Although somewhat primitive in its methodology,
the Badia study offered evidence that olfactory processing is not obstructed during
sleep in humans. That the trigeminal properties of the peppermint odor could have
influenced these findings was not considered, but has since been addressed (Stuck
et al., 2006, 2007).
More recent work in humans compared arousal thresholds of odors administered
during sleep to arousal thresholds of auditory tones (Carskadon and Herz, 2004)
(Fig. 3A and B). Two odors, peppermint and pyridine (fishy, pungent), were
FIGURE 3
Mixed physiological effects of odors on sleeping humans. (A) Delivery of peppermint odor at increasing concentration (one, most dilute; two, least
dilute) had virtually no effect of self-reported behavioral arousals during stage 2, SWS, or REM. In contrast, delivery of auditory tones was highly
successful at eliciting arousals in all sleep stages. (B) Delivery of pyridine odor (fishy, pungent) had a comparatively greater effect on arousals
during lighter stages of sleep (stage 2 and REM), though arousals were inconsistent. Again, tones consistently provoked arousals. (C–E) In a
separate study, four different odorants were delivered during sleep. Compared to a preodor baseline, the first postodor breath was associated with
smaller inhalation volume (C) and larger exhalation volume (D), and across six consecutive breaths postodor (E), the inhale/exhale ratio was
significantly different from baseline, progressively declining from the first to the sixth breath. Effects did not differ across odorants.
Data in (A) and (B) adapted and modified from Carskadon and Herz (2004) and data in (C–E) adapted and modified from Arzi et al. (2010).
318
CHAPTER 12 Smell, Sleep, and Memory
delivered at varying concentrations during overnight sleep throughout all sleep
stages, and subjects were instructed to press a button and give a verbal response
if they smelled an odor. Odors were delivered for 15 s or until behavioral arousal.
If odor delivery failed to arouse the sleeping subject, an auditory tone was then delivered for 5 s or until behavioral arousal, and subjects again pressed a button and
gave a verbal response if they heard a tone. EEG activity was also monitored to determine sleep stage and arousal. Although subjects were aroused by 91% of odors
delivered during stage 1 sleep, peppermint odor failed to evoke a behavioral response
in any other sleep stage, and behavioral response to pyridine odor was sporadic. Occasionally, odor-associated EEG activation occurred in the absence of a behavioral
response, though this too was infrequent. In contrast, subjects were aroused by 97%
of tones across all sleep stages. Researchers concluded that, in stark contrast with
audition, and also in contrast to the Badia study, olfactory responsiveness in all
but the lightest stages of sleep is low to absent. One important limitation of this study
was that data were based on only six participants.
Noting that the trigeminal properties of peppermint and pyridine odorants may
have complicated interpretation of the Carskadon and Herz data, Stuck and colleagues
used a pure olfactory stimulus (H2S) and a pure trigeminal stimulus (CO2), both at
varying concentrations, to disentangle the abilities of olfactory and trigeminal stimuli
to evoke arousal during sleep (Stuck et al., 2007). Arousal was assessed by monitoring
EEG activity only, rather than employing additional behavioral measures. Results
indicated that pure olfactory stimuli had no impact on the frequency of arousals when
delivered during any stage of nREM (including stage 1) even at the highest concentrations, whereas trigeminal stimuli increased arousals during nREM in a dosedependent manner. The same pattern of response to olfactory and trigeminal stimuli
was observed in a follow-up study in which stimuli were delivered during REM
(Grupp et al., 2008).
These studies pointed toward the idea that odors, especially pure olfactants, do
not typically disrupt EEG sleep rhythms in the human brain. However, the inability
of odors to arouse the sleeper does not exclude the possibility that more subtle responses to olfactory stimuli occur during sleep and might be identified using more
precise methods and instruments. Moreover, it is important to note that surface EEG
recordings are notoriously insensitive to signal sources emanating from deep brain
regions within the medial temporal lobe, so it remains possible that electrophysiological arousals in olfactory cortical areas may have eluded detection. Interestingly,
a much earlier study revealed that olfactory stimuli can exert a marked impact on
sleep, whereby sardine odor (a classic feline-relevant smell) delivered during sleep
desynchronized neocortical slow waves in cats (Hernández-Peón et al., 1960).
Research has shown that human subjects make deeper (shallower) sniffs in the
presence of pleasant (unpleasant) odors, with similar respiratory profiles emerging
when subjects merely imagine pleasant or unpleasant smells (Bensafi et al., 2003;
Johnson et al., 2003). Extending this concept, Arzi and colleagues assessed the ability of odors to modify respiratory patterns when delivered during sleep (Arzi et al.,
2010). In this study, four olfactory stimuli were used: two pure olfactory odors
4 The Neurobiological Interface Between Smells and Sleep: Animal Studies
(vanillin, pleasant and ammonium sulfide, unpleasant) and two mildly trigeminal
odors (lavender and vetiver oil). Subjects were exposed to one of the four odors during overnight sleep throughout all sleep stages except for stage 1. Inhalation, exhalation, and the ratio of inhalation to exhalation were measured for each of six breaths
following odor onset, and these were compared to a baseline average of 30 breaths
preceding odor onset. Results showed that onset of each odorant was associated with
decreased inhalation and increased exhalation, with a significant change of the inhale/exhale ratio that was maximal for the first breath and persisted across all six
breaths (Fig. 3C–E). Simultaneous surface EEG recordings suggested that odorants
had no effect on arousals, in line with other studies (Arzi et al., 2012; Carskadon and
Herz, 2004); if anything, odorants reduced the frequency of arousals compared to
baseline. This study was critical in that it helped provide definitive support for
the idea that the odors can modify physiology (in this case respiration) during sleep.
4 THE NEUROBIOLOGICAL INTERFACE BETWEEN SMELLS
AND SLEEP: ANIMAL STUDIES
Animal research on the interaction between sleep and olfactory processing can be
traced back to the French physiologist, Frédérick Bremer, whose pioneering studies
in the 1930s had a far-reaching impact on the the field of neurobiology of sleep
and wakefulness (Siegel, 2002). To elucidate the origin of sleep in the nervous
system, Bremer developed an “isolated brain” (“cerveau isolé”) preparation in cats
(Fig. 4), in which the cerebrum was transected from the rest of the neuraxis at
the level of the rostral midbrain (Bremer, 1935, 1936). This procedure prevented
most sensory inputs from reaching the brain, sparing only the olfactory and optic
nerves. Remarkably, as long as the cat was fortified with artificial ventilation and
a robust cerebrovascular blood supply, it (the cat) persisted in a functional and
FIGURE 4
This mid-sagittal drawing of the cat brain (left) depicts Bremer’s “cerveau isolé” preparation,
with “S” denoting the surgical line of transection between the rostral midbrain and the
lower thalamus, leaving the feline subject (right) in an irreversible sleep-like state and with
pinpoint (miotic) pupils.
Adapted from Bremer (1936).
319
320
CHAPTER 12 Smell, Sleep, and Memory
electrophysiological state closely resembling natural sleep. Inhalation of acetone (an
odor with an admittedly strong trigeminal component) failed to rouse the animal
from its sleep-like state. Bremer concluded that (a) ongoing sensory stimulation must
be essential to maintain a waking state, (b) sleep was the consequence of blocking all
sensory impulses to the cortex, and (c) olfactory and retinal stimulation was insufficient to restore wakefulness. From this work emerged the early and pervasive concept of sleep as a passive state, arising from functional deafferentation of sensory
inputs—wholly incompatible with the idea that sleep might serve an active role in
biological or behavioral processes, let alone that odor inputs might actively modulate
those processes if delivered during sleep.
Actually, it took almost 20 years to show that Bremer’s conclusions about sleep were
incorrect. As pointed out by Arduini and Moruzzi (1953), Bremer had not actually tested
whether odors could disrupt the intrinsic EEG rhythms of the cerveau isolé cat. These
investigators found that gomenol (camphorous odor) and terpineol (floral odor), and
especially forcible introduction of air into the cat’s nostril, were capable of inducing
fast, high-voltage activity in the olfactory bulb (OB), consistent with what Adrian
had shown in urethane-anesthetized animals (Adrian, 1950). Moreover, odor or air stimulation caused a desynchronization of intrinsic EEG slow waves in sensorimotor cortex,
temporal cortex, and thalamus, providing strong evidence that cortical arousal in the cerveau isolé cat could be achieved via external olfactory inputs. As discussed later in this
section, the novel idea that odors had access to the cerebral cortex during sleep-like states
conflicts with some (but not all) of the more recent published animal data.
4.1 Anesthestic Preparations as Models of Sleep-Like States
Apart from sporadic reports in cat (Hernández-Peón et al., 1960), dog (Domino and
Ueki, 1959), rabbit (Faure and Vincent, 1964), and rat (Araki et al., 1980; Gervais
and Pager, 1979, 1982), mechanistic studies linking smells and sleep have gained
momentum only recently. An important set of studies by Fontanini, Bower, and colleagues has challenged the general notion that slow-wave cortical oscillations necessarily reflect a condition where the brain is shut down to external sensory inputs.
These investigators demonstrated that in freely breathing rats anesthetized with a
combination of ketamine–xylazine, slow (<1.5 Hz) field potential oscillations
within both piriform cortex and OB were synchronized with the inspiratory phase
of respiration (Fig. 5), as well as with periodic fluctuations in the piriform cortical
membrane potential (Fontanini et al., 2003). Moreover, in tracheotomized rats—in
which nasal airflow is abolished while breathing persists—the entrainment between
olfactory oscillatory activity and respiratory rhythm was disrupted, suggesting that
the tactile sensation of air movement through the nose (rather than breathing per se)
is critical for synchronizing brain and behavioral states (Fig. 5B–D). Such results
imply that olfactory brain areas remain receptive to external afferent input during
slow-wave states, unlike sensory neocortical areas in which sensory input appears
to be blunted. An important take-home message (which will be a main focus of
the next section) is that odor stimuli may have privileged access to the central nervous system during SWS.
4 The Neurobiological Interface Between Smells and Sleep: Animal Studies
FIGURE 5
Persistence of respiratory-driven activity during sleep-like states. (A) In freely breathing rats in
a deep sleep-like state under ketamine–xylazine anesthesia, membrane potential oscillations
in piriform cortex (Vm), and local field potentials (LFPs) in piriform cortex and OB are
entrained to respiration (downward deflection represents inhalation). (B) In tracheotomized
rats, the respiratory coupling to cortical and bulbar oscillations is disrupted, reflecting the
importance of airflow through the nose to entrain these rhythms. (C) Cross-covariance
analysis indicates that in intact rats (dashed line), there is a positive correlation between Vm
and LFPs in OB that peaks near time 0, whereas in tracheotomized rats (solid line), this phase
relationship is inverted. (D) The strong in-phase relationship between respiration and Vm
observed in intact rats (dashed line, peaking at 225–315 ) is virtually absent in
tracheotomized rats (solid line). Start of inhalation is at 0 .
Adapted from Fontanini et al. (2003).
In follow-up work, Fontanini and Bower (2005) also found that by varying the
depth of anesthesia, they could modulate functional coupling within the rodent olfactory system. Under deep anesthesia, slow oscillatory responses (0.5–1.5 Hz) in
OB and piriform cortex were strongly correlated to each other and to respiration,
whereas under light anesthesia, fast oscillatory responses (>15 Hz) in these two regions became uncoupled, and only bulb activity remained in phase with the
321
322
CHAPTER 12 Smell, Sleep, and Memory
respiratory cycle. Insofar as these slow and fast electrical profiles correspond to
states of SWS and REM/waking, respectively, the data further reinforce the idea that
olfactory cortical regions are not isolated from the sensory periphery during sleep,
but can be engaged with entry of air into the nose. Whether such effects are mediated
by nonspecific activation of olfactory sensory neurons (Grosmaitre et al., 2009), by
intranasal chemoreceptors that detect airflow-induced pressure changes (Grosmaitre
et al., 2007), or by cortical (or subcortical) centrifugal projections remains unclear.
While the above studies examined olfactory sleep-like states in the absence of
odor stimulation, Murakami and colleagues went one step further by testing whether
odor-evoked responses in OB and cortex were state dependent (Murakami et al.,
2005). To this end, rats were anesthetized with urethane, an agent that has the unique
property of generating spontaneous alternations of slow-wave and fast-wave neocortical EEG patterns. Delivery of odor elicited significantly greater single-unit activity
in anterior piriform cortex (APC) during fast-wave states than during slow-wave
states (Fig. 6A), an effect that was seen during both natural and artificial respiration,
FIGURE 6
Sleep-dependent sensory “gating” of odor inputs between rodent OB and APC. (A) In
urethane-anesthetized rats, odorant evoked significantly lower single-unit activity in the APC
during slow-wave states (SWS, left) than during fast-wave states (FWS, right). (B) In contrast,
odorant-evoked activity was similarly robust during both SWS and FWS in OB. Note that neither
single-unit responses nor EEG oscillations appear to be in synchrony with artificially paced
respiration of the animal. EEG recordings to monitor SWS and FWS were obtained from occipital
neocortex.
Reproduced from Murakami et al. (2005), with permission from Elsevier, # 2005.
4 The Neurobiological Interface Between Smells and Sleep: Animal Studies
though spike responses were not coupled to respiration in either EEG state. In contrast to the dampening of odor-evoked responses in APC during the slow-wave state,
odor-evoked activity in OB persisted during both fast-wave and slow-wave states
(Fig. 6B), and electrical stimulation of OB elicited significantly fewer spike responses in APC during slow-wave (vs. fast-wave) states. These data were interpreted
as evidence for sleep-dependent gating of odor processing at the level of piriform
cortex. Given that there is no requisite thalamic intermediary between the olfactory
periphery and olfactory cortex, the authors also conjectured that piriform cortex
might in fact fulfill the type of gating function ascribed to the thalamus in other sensory systems. However, because single-unit recordings were confined to OB and
olfactory cortex, it remains possible that gating is imposed via top-down projections
from higher order regions, such as mediodorsal thalamus or orbitofrontal cortex.
4.2 Olfactory Cortical Processing During Natural Sleep States
The above findings in anesthetized preparations have been more recently extended
to freely behaving rats during natural sleep (Manabe et al., 2011). Extracellular
recordings from APC during SWS revealed spontaneous irregular slow oscillations
(0.5–4 Hz) of local field potentials (LFPs), as well as olfactory cortical sharp
waves accompanied by synchronized spike firing (Fig. 7A). Sharp waves were also
identified in OB and were shown to be in synchrony with sharp waves in APC. The
spontaneous manifestation of APC sharp waves was markedly suppressed during the
wake state and during REM (Fig. 7B). Taking these results together with data from
the Murakami study, the investigators conceived an olfactory framework of SWS in
which the afferent flow of information is blunted (gated) at the level of OB, and the
associative flow of information between higher order areas, olfactory cortex, and OB
is heightened—effectively a “turning inward” of information processing to promote
memory consolidation, with olfactory sharp waves mediating a replay of odor events
experienced in the waking state. The general idea that SWS offers protection from
external interference in order to promote memory replay and consolidation is a dominant theme in the current sleep literature (Rasch and Born, 2013) that will be further
addressed in Section 5 of this chapter.
4.3 Impact of Odor Experience on Olfactory Processing and Behavior
Recent research from the Wilson lab has framed the question of smell and sleep in a
different way: if SWS presents an optimal window in which memories can be reinforced for events (and odors) experienced during waking, then experimental manipulations designed to enhance odor experience should elicit neural changes in
olfactory areas of the sleeping brain. In one of these studies, urethane anesthesia
was used to evoke spontaneous fast-wave and slow-wave EEG states (as discussed
earlier), during which single-unit activity in rodent piriform cortex was recorded
(Wilson, 2010). First, spontaneous cortical activity was measured during slow-wave
activity, establishing a baseline of spike firing in APC (Fig. 8A). Then, odor stimuli
were delivered to the animal during fast-wave activity, mimicking odor experience
323
324
CHAPTER 12 Smell, Sleep, and Memory
FIGURE 7
Spontaneous generation of olfactory sharp waves during natural sleep. (A) LFP recordings in
rodent APC during SWS demonstrate the irregular emergence of large-amplitude sharp waves
(downward deflections; arrows) that are aligned with phasic bursts of multiunit APC activity. (B)
Both the frequency (top) and amplitude (middle) of the olfactory cortical sharp waves are
increased during SWS and are virtually absent during other behavioral epochs, including
exploration, grooming, and REM (bottom). The time-frequency power spectrum is based on EEG
activity recorded from occipital cortex.
Adapted from Manabe et al. (2011).
that would occur in a wake-like state. Critically, a wide panel of odorants was tested
during this period, enabling identification of odors that were able to drive activity of
the neurons being recorded. These “adequate” odors were then delivered repeatedly
during the fast-wave state. Finally, when the animal transitioned back to the slowwave state, spontaneous firing in APC was again monitored, this time in the absence
of odor.
By aligning single-unit activity to the peak of the LFP slow-wave, Wilson found
that odor experience significantly modulated spike firing profiles for cells in APC, in
comparison to a control group that did not receive odor (Wilson, 2010). Whereas
4 The Neurobiological Interface Between Smells and Sleep: Animal Studies
FIGURE 8
Effect of odor experience on olfactory cortical activity in sleep. (A) In urethane-anesthetized
rats, either odor or odorless control was delivered during fast-wave activity (FWA).
Before and after olfactory presentation, spontaneous single-unit activity during slow-wave
activity (SWA) was recorded in APC pyramidal layers II/III. (B) In control rats, there was
no significant change in the temporal profile of spike firing from preodor SWA to postodor
SWA when unit activity was aligned to the slow-wave peak (black waveform, top). (C) In the
odor-experienced rats, there was a significant shift in the temporal structure of the response,
with some cells firing earlier and some firing later, with respect to the slow wave. (D) An
example from one APC neuron demonstrates a shift in peak unit activity following odor
exposure, in this case from earlier to later in the slow-wave period.
Modified and adapted from Wilson (2010).
spike firing remained tightly entrained to the phase of slow waves in the control
group, the temporal structure of spike firing in the odor group became destabilized
from pre- to postodor stimulation (Fig. 8B–D). There was also a significant change in
the variability of the leading (early) edge of the slow-wave LFP from pre- to postodor, suggesting an impact on the population-level response. Based on these findings, it is tempting to conclude that the experience-dependent neural changes in
the slow-wave state might reflect memory reactivation or replay of odor events
325
326
CHAPTER 12 Smell, Sleep, and Memory
occurring in the preceding fast-wave state, though the absence of corresponding
behavioral data makes it difficult to substantiate these claims.
Complementary experiments from the Wilson lab have extended our understanding
of the functional impact of sleep-like states on olfactory network plasticity. In rats under urethane anesthesia, the coherence of spontaneous spike firing activity between
APC and OB (within delta and theta bands) decreased from the fast-wave state to
the slow-wave state, whereas coherence between APC and hippocampus, and between
APC and basolateral amygdala, significantly increased (Wilson and Yan, 2010)
(Fig. 9). In addition, Granger causality analysis suggested that information flow from
hippocampus to APC was stronger during slow-wave (vs. fast-wave) activity, and was
also significantly stronger than in the reverse (APC to hippocampus) direction. In a separate study, functional magnetic resonance imaging (fMRI) resting-state connectivity
(reflecting very low-frequency oscillations <0.1 Hz) was measured in urethaneanesthetized rats (Wilson et al., 2011). Because EEG measures could not be obtained
during fMRI scanning, fast- and slow-wave states were estimated on the basis of differences in breathing rates. Using this indirect index, Wilson and colleagues identified
FIGURE 9
Sleep-induced modulation of olfactory network connectivity. Spontaneous LFP activity
was recorded during urethane transitions between fast-wave activity (FWA) and slow-wave
activity (SWA), with electrodes placed in rodent OB, layers I and III of anterior piriform
cortex (PCX), and dorsal hippocampus (dHPC). Mean waveform coherence within both the
delta (A) and theta (B) bands decreased between PCX III and OB from FWA to SWA, and
increased between PCX III and dHPC from FWA to SWA.
Adapted from Wilson and Yan (2010).
5 Olfactory Cues can Modulate Cognitive Processes During Sleep
significant state-dependent changes from fast-wave to slow-wave states, including enhanced functional connectivity between piriform cortex and dorsal hippocampus, piriform cortex and neocortex, and dorsal hippocampus and amygdala. Both this study and
the Wilson and Yan study (Wilson and Yan, 2010) illustrate the dynamic effects of
sleep-like state-dependent shifts in olfactory network interactions with both limbic
and neocortical systems, which could help promote memory consolidation.
Given the uncertainties regarding the validity of anesthetized states to model natural sleep, Barnes and colleagues investigated odor-evoked piriform activity in unanesthetized, chronically recorded rats (Barnes et al., 2011). Compared to wake and
REM states, SWS was associated with reduced odor-evoked oscillatory activity in
APC across theta, beta, and gamma bands, confirming a relative lack of olfactory responsiveness during SWS. Another goal of this study was to test the effect of experience—in this case, olfactory fear conditioning—on SWS. After learning to associate
odors with foot shock in the awake state, rats spent more time in SWS compared to
controls, and interestingly, on a subject-by-subject basis, the amount of time spent
in SWS correlated with the strength of conditioned fear as indexed by freezing duration. Put differently, the extra time spent in SWS facilitated memory consolidation,
possibly by extending the period of reduced external influence and enhancing cortical
associative interactions.
5 OLFACTORY CUES CAN MODULATE COGNITIVE PROCESSES
DURING SLEEP
Over the past 20 years, it has become increasingly clear that odors have a profound impact on the sleeper despite their inability to consistently evoke arousal. These qualities
make odors ideal candidates to intervene during sleep without disrupting it. Notably, two
distinguishing anatomical features of the olfactory system may indicate a privileged role
for odors in memory modulation in the sleeping brain. First, the olfactory system shares
considerable overlap with limbic and paralimbic brain areas that mediate emotion,
memory, and behavior (Gottfried, 2006), likely reflecting the fact that, throughout much
of the animal kingdom, the sense of smell has been paramount for survival. Second, olfactory sensory neurons in the nasal epithelium project to the OB, which in turn projects
directly to piriform cortex, and from there to many downstream cortical regions. The
important point is that odor information can be relayed to the cortex without having
to pass through the thalamus, which in other sensory systems is thought to serve a “gating” function that impedes cortical transmission of sensory information during sleep.
5.1 Targeting Declarative Memories During Sleep
Over the past decade, researchers have employed olfactory cues in order to modulate
memory consolidation, and even to demonstrate that learning can occur during sleep.
In a pivotal study, Rasch and colleagues showed that odors could facilitate memory
consolidation while human subjects were asleep (Rasch et al., 2007). Subjects performed a visuospatial object–location task in which they were required to learn the
327
328
CHAPTER 12 Smell, Sleep, and Memory
unique locations of 15 visual stimuli, in the presence of either a rose odor or an odorless
vehicle (control group) (Fig. 10A). During subsequent sleep, the rose odor was delivered
in alternating on/off blocks of 30 s, to minimize odor habituation. Remarkably, when
the odor was delivered during SWS, subjects in the odor group remembered the picture
locations more accurately the following day than did subjects in the control group
(Fig. 10B). This relative memory boost was not observed if the odor was delivered only
during SWS, or if odor delivery following learning occurred during REM or during a
wake state (Fig. 10C–E). In a complementary experiment, subjects learned the same
visuospatial task in the presence of rose odor, then underwent fMRI scanning while
the same odor was delivered during either SWS or wake. Odor-evoked hippocampal
activity increased in the SWS group compared to the wake group, perhaps suggesting
that presentation of a relevant odor context in sleep promoted reactivation of the associated hippocampus-dependent visuospatial memories, with consequent gains in memory performance the next day. This technique of using sensory cues to selectively
modulate memory consolidation has also proven effective with auditory cues (Antony
et al., 2012; Fuentemilla et al., 2013; Oudiette et al., 2013a; Rudoy et al., 2009) and
has been termed targeted memory reactivation (TMR) (Oudiette and Paller, 2013).
In follow-up studies from the same group, subjects again took part in the aforementioned visuospatial memory task in the presence of a background odorant,
isobutyraldehyde (“sweaty, sour” odor) (Diekelmann et al., 2011). The difference
here was that after the sleep phase (or an awake control phase), subjects learned
another set of object–location pairs, as a way to test whether sleep could shield
the original (presleep) memories from the new interfering (postsleep) memories.
Results showed that delivery of the olfactory cue during subsequent SWS stabilized
visuospatial memories against interference, whereas the same cue delivered during
the wake state rendered those memories more vulnerable to interference. A related
fMRI experiment from this same study (Diekelmann et al., 2011) revealed that presentation of a sleep-associated odor during wake elicited activation in lateral prefrontal cortex, when compared to a control presentation of the same odor that had not
been delivered during sleep. Again, significant odor responses in SWS were observed in hippocampus as well as retrosplenial cortex. These researchers have also
found that delivering an olfactory cue during a 40-min nap facilitated memory consolidation to the same extent as did an uncued 90-min nap, whereas an uncued 40min nap did not significantly benefit memory (Diekelmann et al., 2012). That memory performance following the 40-min odor-cued nap and the 90-min uncued nap
were equivalent raises questions regarding the general robustness of the memory
effects, and at a minimum illustrates the critical dependence of these effects on task
parameters and timing.
5.2 Targeting Emotional Memories During Sleep
Though TMR has been employed most frequently to enhance declarative memories,
the technique has also proven effective in modulating other types of memories. In
recent work from our lab, Hauner and colleagues were able to extinguish fear
5 Olfactory Cues can Modulate Cognitive Processes During Sleep
FIGURE 10
The first targeted memory reactivation study using olfactory cues to enhance declarative
memory. (A) In a learning session taking place in the evening, subjects learned a visuospatial
object–location task in the presence of rose odor. In subsequent sleep, rose odor or an
odorless vehicle (control) was delivered during SWS. Memory for the object locations was
assessed the following morning in the absence of odor. (B) Presentation of odor during both
learning and SWS enhanced postsleep recall for the object (card) locations, compared to the
control condition where odor was presented during learning only. (C) There was no effect of
odor delivery on recall when odor was delivered during SWS but not during prior learning.
There was also no effect on memory retrieval when odor was delivered during learning and
again during subsequent REM (D) or subsequent wake (E).
Reproduced from Rasch et al. (2007), with permission from AAAS.
memories using olfactory TMR during SWS in human subjects (Hauner et al.,
2013). In an initial fMRI conditioning session (Fig. 11A), subjects viewed two
faces in the context of a “target” (tg) odor background. One face (the conditioned
stimulus, or tgCS þ) was paired on 50% of trials with a foot shock; the other face
was not paired with shock (the control stimulus, or tgCS ). As a control for the tg
odor context, in separate conditioning blocks, another two faces were presented in
the presence of a “nontarget” (nt) odor context, resulting in ntCS þ and ntCS 329
330
CHAPTER 12 Smell, Sleep, and Memory
FIGURE 11
Experimental paradigm of the fear conditioning task. (A) Olfactory contextual fear conditioning
was completed in the MRI scanner. “Target” (blue) and “nontarget” (green) odorants
served as background contexts in alternating trial blocks. Two conditioned stimuli (face images)
were presented within the target context (tgCSþ, tgCS), and two within the nontarget
context (ntCSþ, ntCS). Both tgCSþ and ntCSþ were paired with the US (mild electric shock)
on 50% of trials (tgCS þ p, ntCS þ p) and were unpaired on the remaining 50% of trials
(tgCS þ u, ntCS þ u). Control stimuli (tgCS, ntCS) were never paired with shock. Stimuli
were presented in eight blocks (four target, four nontarget), in pseudo-randomized order.
(B) Subjects in SWS underwent repeated reexposure to the target odorant (in 30-s on/off
intervals), outside of the MRI scanner. The hypnogram illustrates sleep-staging data for
a representative subject. (C) Upon waking, subjects completed a retrieval task in the MRI
scanner. This task was identical to the presleep conditioning task, apart from a 12.5%
partial-reinforcement schedule to prevent extinction.
Modified from Hauner et al. (2013).
5 Olfactory Cues can Modulate Cognitive Processes During Sleep
stimuli. Skin conductance responses (SCRs) were measured as a physiological index of fear conditioning.
In a subsequent session outside the fMRI scanner, each subject took a 90-min nap
during which the tg odor was delivered in alternating 30 s on/30 s off blocks during
SWS (Fig. 11B). The goal here was to modulate fear memories selectively for the
conditioned stimulus that had been previously encountered in the presence of the
tg odor (i.e., tgCS þ). During the first half of SWS, the tg odor evoked a robust
SCR, compared to odor-off intervals, but during the second half of SWS, this response decreased significantly. These findings suggest that reexposure to the tg odor
context in sleep induced within-session fear extinction. Finally, after waking, subjects returned to the fMRI scanner and took part in the same fear conditioning task
(Fig. 11C), for direct comparison to the presleep session.
Analysis of the odor-evoked SCR data revealed a significant decrease in the
“fear” response from pre- to postsleep for the tgCS þ face versus the ntCS þ face,
and in comparison to the nontarget conditions (Fig. 12A). These findings suggest that
delivery of the tg odor during SWS promoted the emergence of fear extinction in the
FIGURE 12
Effects of TMR on conditioned fear response in humans. (A) SCR from pre- to postsleep
decreased for tgCS þ compared to ntCS þ. (B) Duration of odor delivery during SWS was
correlated with degree of SCR reduction from pre- to postsleep. (C) Anterior hippocampal
activity in response to tgCS þ decreased from pre- to postsleep compared to ntCS þ.
(D) Voxel-wise ensemble maps of left amygdala activity in one subject (left) illustrate that,
from pre- to postsleep, response patterns were more decorrelated when elicited by
tgCS þ compared to ntCS þ.
Modified from Hauner et al. (2013).
331
332
CHAPTER 12 Smell, Sleep, and Memory
postsleep state. Interestingly, the degree to which the fear response decreased was
correlated with the duration of tg odor delivery during SWS within individual subjects (Fig. 12B). These behavioral effects were also found to differ significantly from
an independent group of wake subjects. The implication is that sleep constitutes a
unique state in which targeted fear memories can be selectively extinguished, while
nontargeted fear memories remain intact.
Concurrent analysis of the fMRI data showed that the cue-evoked mean activity
in anterior hippocampus decreased from pre- to postsleep selectively for tgCS þ versus ntCS þ (adjusted for CS baselines) (Fig. 12C). The reduction of activity in the
hippocampus is consistent with other work suggesting that sleep accelerates consolidation of nonemotional memories, with reduced hippocampal dependence as learning proceeds (Wang et al., 2009). Interestingly, although there was no direct measure
of memory for the tgCS þ and ntCS þ faces, we did find that reaction times (as an
indirect measure of stimulus recognition) were significantly faster in response to
tgCS þ than to ntCS þ. Taken together, it is reasonable to speculate that delivery
of the tg odor in SWS resulted in enhanced recognition memory for the tgCS þ,
reduced hippocampal engagement, and weakened behavioral expression of fear.
Finally, to the extent that presentation of the tg odor during SWS induced a fundamental shift in the “meaning” of the tgCSþ (i.e., from a conditioned fear memory in
presleep to an extinguished fear memory in postsleep), we used fMRI multivariate techniques to test whether cue-evoked fMRI ensemble patterns in the amygdala significantly
differed as a result of odor delivery during sleep. Data showed that the fMRI patterns
evoked by tgCSþ significantly diverged (became more decorrelated) in the amygdala
from pre- to postsleep, compared to the ntCSþ (Fig. 12D and E). This reorganization of
amygdala ensemble activity suggests that odor contextual reexposure in SWS induces
formation of a qualitatively unique memory trace in the amygdala (in line with findings
in animal models; Herry et al., 2008), rather than weakening or “erasure” of the original
trace per se.
Interestingly, a somewhat analogous experiment in mice reached an opposite
conclusion (Rolls et al., 2013). Mice were conditioned to associate an odor with
the delivery of a foot shock. After a 24-h delay, either the conditioned odor (CS)
or a control odor was delivered during nREM. After another 24-h delay, the odorevoked fear response was assessed by measuring the duration of freezing behavior
following odor delivery (Fig. 13A). Although there were no significant differences in
EEG delta power during delivery of CS odor versus control odor in sleep (Fig. 13B),
in the postsleep behavioral assessment, mice in the conditioned odor group demonstrated increased fear response to the CS odor, compared to mice receiving the
control odor (Fig. 13C). When a protein synthesis inhibitor was injected into the
basolateral amygdala prior to odor reexposure, the opposite trend was observed. That
is, the fear response following odor reexposure was attenuated rather than strengthened. Although these results seem to contradict those from the Hauner et al. study,
there are several differences in the experimental designs that may account for the
incongruous results (Oudiette et al., 2013b). Most notably, the 24-h delay period
prior to odor reexposure allotted in the Rolls study likely led to more stable memory
5 Olfactory Cues can Modulate Cognitive Processes During Sleep
FIGURE 13
Effects of TMR on conditioned fear response in mice. (A) Fear-conditioning paradigm in which
10-s delivery of amyl acetate (CS) predicted a 10-s foot shock. After a 24-h interval, CS or
control odor was delivered during nREM. After an additional 24-h interval, freezing behavior
in response to CS was assessed in a novel environment as an index of conditioned fear.
(B) Spectral analysis revealed no differences in delta activity between CS and control groups
during nREM. (C) Odor-evoked freezing duration increased for the CS group compared to
controls.
Adapted and modified from Rolls et al. (2013).
storage than in the Hauner study where odor reexposure occurred within a couple of
hours after conditioning. Another critical difference is that the odor served as a context cue in the Hauner study, as opposed to standing in for the CS itself in the Rolls
study. Taken together, these studies illustrate the ability of olfactory cues to
strengthen or attenuate emotional memories depending on experimental conditions.
This is an important finding, as it may implicate TMR as a potential treatment for
psychological disorders, such as posttraumatic stress disorder, which could prove
to be a less stressful alternative to current reexposure treatments.
5.3 De Novo Olfactory Learning in Sleep
In spite of what pop culture might have taught us, placing a textbook underneath your
pillow while asleep does not generally lead to higher test scores the next morning.
More systematic efforts to demonstrate the ability to learn new material during sleep
have been attempted. As early as 1942, Leshan tried to break a group of summer
camp attendees of a nail-biting habit by delivering the message “my fingernails taste
terribly bitter” via portable electric phonograph 300 times per night for 54 nights
333
334
CHAPTER 12 Smell, Sleep, and Memory
(Leshan, 1942). Of 40 children, 8 of 20 in the experimental group stopped biting their
nails (vs. 0 of 20 controls). As there was no physiological measure to ensure that the
children were actually sleeping when the message was delivered (not to mention the
likely lack of an IRB protocol), results from this study are far from convincing. However, Leshan’s efforts epitomize the long-standing interest in learning during sleep.
Given the apparent unique access of odor inputs to the sleeping brain, Arzi and
colleagues set out to test this question using an associative conditioning paradigm
(Arzi et al., 2012). This study was motivated by two prior findings: first, that odors
delivered in sleep can in fact modulate breathing patterns (Arzi et al., 2010; see
Section 3 above); and second, that subjects make larger inhalations in the presence
of pleasant odors than unpleasant odors (Bensafi et al., 2003). The basic idea was to
determine whether a tone paired with a pleasant (vs. unpleasant) odor during sleep
would itself evoke larger (vs. smaller) inhalations in the absence of odor. While subjects were asleep, one of two distinct tones was delivered every 25–45 s. On twothirds of trials, tones were followed by a unique odor after 2.7 s; on the remaining
third of trials, tones were unpaired (Fig. 14A). This paradigm conformed to a partialreinforcement trace conditioning procedure and provided a way to dissociate tone
effects from odor effects. Importantly, one tone was consistently paired with a pleasant odor, while the other was consistently paired with an unpleasant odor. Both EEG
and respiration were monitored throughout. The next morning when subjects awoke,
the two tones were presented again (without odor), alongside a third novel tone
(Fig. 14B), to determine the strength of memory retention, again using respiratory
measures.
While subjects slept in blithe ignorance during tone-odor conditioning, the volume of their odor-evoked “sniffs” was found to be larger for the pleasant odor than
for the unpleasant odor (Fig. 14C, left), as expected based on prior work (Arzi et al.,
2010). Critically, as conditioning proceeded during sleep, inhalation volumes became larger in response to the tone that had been paired with pleasant odor, compared
to the tone that had been paired with unpleasant odor (Fig. 14C, middle). Moreover,
these effects persisted into the next morning in the waking state: pleasant odorassociated tones evoked larger breaths than did unpleasant odor-associated tones
(Fig. 14C, right). The effects observed during wake were less pronounced than those
observed during preceding sleep, and when analysis was restricted to subjects that
did have a single arousal within 30 s of tone onset during conditioning, responses
to conditioned tones did not differ significantly from the response to the novel tone.
Nonetheless, these exciting findings provide unique evidence that new information
can be learned during sleep.
6 FORMS OF OLFACTORY HOMEOSTASIS DURING SLEEP
A handful of studies have considered the homeostatic effects of sleep on olfactory
system processing at both the metabolic and cellular levels. Based on the observation
that postprandial states are often followed by drowsiness and sleep, Gervais and
6 Forms of Olfactory Homeostasis During Sleep
FIGURE 14
Effects of tone–odor trace conditioning during sleep on sniff response. (A) During sleep, one
tone predicted delivery of pleasant odor and a second tone predicted delivery of unpleasant
odor (67% partial reinforcement). Sniff volume evoked by tones alone was measured as an
index of conditioning. (B) During wake, sniff volume evoked by conditioned tones and an
additional control tone was measured to assess efficacy of conditioning. (C) Sniff volume
increased in response to pleasant smells compared to unpleasant smells during sleep (left).
Sniff volume increased in response to pleasant odor-associated tone compared to unpleasant
odor-associated tone during sleep and during subsequent wake (middle and right).
Adapted and modified from Arzi et al. (2012).
Pager (1979) reasoned that nutritional status and food odors might have a direct
impact on sleep physiology. Multiunit recordings from the rodent OB showed that
food odor increased mitral cell activity in awake hungry rats, compared to a nonfood
odor, and compared to awake satiated rats. In contrast, during SWS, mitral cell activity increased in response to food versus nonfood odor, but there was no difference
between hunger and satiety states. Moreover, while food odor delivered during SWS
elicited an increased number of neocortical arousals (on EEG) in the hungry state
compared to the satiated state, this effect was also evoked by nonfood odors, suggesting that hunger induces a nonspecific arousal response to olfactory stimuli during
sleep. In subsequent work (Gervais and Pager, 1982), the ability of food odor to elicit
a higher rate of neocortical desynchronization in the hungry (vs. satiated) state during
335
336
CHAPTER 12 Smell, Sleep, and Memory
SWS was abolished in rats with a lesion of the medial olfactory tract (i.e., anterior
limb of the anterior commissure). The satiety-dependent effects were preserved
in another group of rats with a lesion of the lateral olfactory tract. These results
imply that behaviorally relevant olfactory information (in this case, food odor in
food-deprived rats) can still access the brain during SWS, specifically via medial
projections from OB that join the medial forebrain bundle en route to the reticular
formation. Such mechanisms might ensure that in the case of starvation, a sleeping
rat would not miss a meal should the smell of food waft into its bed chamber.
Curiously, interest in the effects of postprandial states on the olfactory system
during sleep has been recently revived. Based on the idea that adult neurogenesis
in OB is regulated by sensory experience, Yokoyama and colleagues tested whether
food experience could influence the survival of adult-born granule cells (GCs) in the
postprandial period (Yokoyama et al., 2011). When placed on a restricted feeding
schedule, mice engaged in a range of behaviors following food consumption, including grooming, resting, and sleeping. In parallel, the number of caspase-3-activated
GCs (where caspase-3 is a marker of apoptotic cell death) increased during this postprandial period, but only between 1 and 2 h after feeding. Of note, the duration of
postprandial SWS correlated with the number of caspase-3-positive GCs, and if mice
were disturbed or kept awake during this critical window, the effects on GC apoptosis were not identified. Somewhat oddly, these postprandial effects were not demonstrated in mice that were allowed to eat freely. Though this study raises many
questions, the results point toward a role for SWS in reducing the number of
adult-born neurons in OB, as a form of cellular homeostasis and structural reorganization. Given that ongoing olfactory sensory experience promotes an exuberant birth
of many new GCs each day, a mechanism must be in place to eliminate those cells
that are not incorporated effectively into the circuit. Sleep, perhaps with an instructive neuroendocrine or neuromodulatory signal, may help determine which of those
GCs are ultimately destined for survival.
7 CONCLUSIONS
Smells and sleep have both held great relevance for survival since the earliest days of
nervous system evolution. In contrast to some of the other senses, it seems that the
olfactory system has successfully adapted to the behavioral necessities of sleep, providing a conduit between the external world and the internal world of the brain for an
otherwise unresponsive animal. Certainly, there is strong evidence in animal models
that neural oscillations in olfactory cortex remain in synchrony with respiratory
rhythms during SWS (Fontanini and Bower, 2005; Fontanini et al., 2003). Because
the delivery of odor information to the brain necessarily fluctuates with the ebb and
flow of respiration, it makes sense that a cortical state of “readiness” in the olfactory
system should be entrained to breathing. Indeed, as shown in hungry rats, delivery of
a salient food odor during SWS is capable of generating physiological arousals
(Gervais and Pager, 1982) that could improve the chances of securing a meal.
7 Conclusions
Compatible with the idea that there exists a direct line from the outside world to the
olfactory system during sleep, research in sleeping human subjects convincingly
demonstrates that odors delivered during SWS can have a significant impact,
both behavioral and neural, on expression of declarative memory, fear memory,
and associative learning (Arzi et al., 2012; Diekelmann et al., 2011, 2012; Hauner
et al., 2013; Rasch et al., 2007).
In addressing the question of whether the olfactory system remains “open” during
SWS, certain animal studies have arrived at opposite conclusions. Different research
groups have found that odor-evoked single-unit activity in piriform cortex is considerably reduced during sleep-like slow-wave states, compared to sleep-like fast-wave
states (Murakami et al., 2005; Wilson, 2010). These findings are taken to suggest that
the olfactory system is “closed” to external odor stimulation during SWS, perhaps
echoing the same functional organization that exists in the hippocampus where
the “closed” state facilitates memory reactivation in the absence of outside interference (Wang et al., 2009). Although this is an attractive hypothesis, there are several
reasons to adopt a more cautious perspective.
First, just because olfactory brain areas are less responsive to odors during SWS
does not mean that they are unresponsive, a point actually made by Wilson and colleagues (Barnes et al., 2011). Even if one accepts that the flow of odor information
from OB to piriform cortex is totally obstructed, there are still many other potential
avenues by which odor information could engage the brain, since the OB also sends
direct monosynaptic projections to the amygdala and entorhinal cortex, which themselves are linked with orbitofrontal cortex and other limbic and paralimbic regions.
Second, though anesthetics are often used to induce sleep-like states that resemble the behavior and neurophysiology of natural sleep, this does not mean that the
anesthetized brain is a veridical model of the sleeping brain. Indeed, classic work
by Domino and Ueki (1959) demonstrated that a myriad of different anesthetics each
had unique effects on rhinencephalic and neocortical EEG rhythms. Thus, some of
the reported differences in the animal studies described here might be attributable to
this confounding factor. The need to examine natural sleep states in the absence of
anesthesia will be critical for clarifying the functional organization of the olfactory
system in sleep, and recent research has begun to move in this direction (e.g., Barnes
et al., 2011; Manabe et al., 2011).
Finally, because respiration can substantially vary during wake, nREM, and
REM, it is imperative to consider this parameter in sleep-based analyses of
olfactory processing. For example, the demonstration of odor-evoked differences
between sleep stages could simply reflect the fact that odor was delivered less efficiently in one of those stages, rather than reflecting intrinsic neurophysiological differences. Moreover, olfactory studies have long shown that the temporal relationship
between odor onset and onset of inhalation (sniff) has a crucial impact on olfactory
perception and odor coding (Bathellier et al., 2008; Buonviso et al., 2003; Carey
and Wachowiak, 2011; Chaput, 1986; Shusterman et al., 2011), but in some of
the sleep studies reported here, odor-evoked responses were characterized without
regard to respiratory phase, possibly accounting for some of the observed variability.
337
338
CHAPTER 12 Smell, Sleep, and Memory
Throughout history, sleep has been viewed as a gateway to knowing the unknown. It is the stuff of primal emotions: nightmares, night terrors, and nocturnal
emissions. Sleep is the territory of one of the most classic psychological forays into
the human mind (Freud, 1899), and of one of the most critically acclaimed comic
book series of all time, The Sandman (Gaiman, 1991–1996). Sleep has emerged
as an increasing focus of neuroscience research, serving as a powerful gateway to
understanding brain function and behavior. Recent studies have only begun to probe
the interface between sleep and smells, and the future holds great promise for the use
of odor-based manipulations to clarify basic mechanisms of olfactory perception,
and even to elucidate the ever-enigmatic function of sleep.
References
Adrian, E., 1950. The electrical activity of the mammalian olfactory bulb. EEG Clin. Neurophysiol. 2, 377–388.
Antony, J.W., Gobel, E.W., O’hare, J.K., Reber, P.J., Paller, K.A., 2012. Cued memory reactivation during sleep influences skill learning. Nat. Neurosci. 15, 1114–1116.
Araki, H., Yamamoto, T., Watanabe, S., Ueki, S., 1980. Changes in sleep-wakefulness pattern
following bilateral olfactory bulbectomy in rats. Physiol. Behav. 24, 73–78.
Arduini, A., Moruzzi, G., 1953. Olfactory arousal reactions in the “cerveau isolé” cat. Electroencephalogr. Clin. Neurophysiol. 5, 243–250.
Arzi, A., Sobel, N., 2011. Olfactory perception as a compass for olfactory neural maps. Trends
Cogn. Sci. 15, 537–545.
Arzi, A., Sela, L., Green, A., Givaty, G., Dagan, Y., Sobel, N., 2010. The influence of odorants
on respiratory patterns in sleep. Chem. Senses 35, 31–40.
Arzi, A., Shedlesky, L., Ben-Shaul, M., Nasser, K., Oksenberg, A., Hairston, I.S., Sobel, N.,
2012. Humans can learn new information during sleep. Nat. Neurosci. 15, 1460–1465.
Badia, P., Wesensten, N., Lammers, W., Culpepper, J., Harsh, J., 1990. Responsiveness to olfactory stimuli presented in sleep. Physiol. Behav. 48, 87–90.
Barnes, D.C., Chapuis, J., Chaudhury, D., Wilson, D.A., 2011. Odor fear conditioning modifies piriform cortex local field potentials both during conditioning and during postconditioning sleep. PLoS One 6, e18130–e.
Bathellier, B., Buhl, D., Accolla, R., Carleton, A., 2008. Dynamic ensemble odor coding in the
mammalian olfactory bulb: sensory information at different timescales. Neuron 57,
586–598.
Bendor, D., Wilson, M., 2012. Biasing the content of hippocampal replay during sleep. Nat.
Neurosci. 15, 1439–1444.
Bensafi, M., Porter, J., Pouliot, S., Mainland, J., Johnson, B., Zelano, C., Young, N.,
Bremner, E., Aframian, D., Khan, R., Sobel, N., 2003. Olfactomotor activity during imagery mimics that during perception. Nat. Neurosci. 6, 1142–1144.
Besedovsky, L., Lange, T., Born, J., 2012. Sleep and immune function. Pflüg. Arch. Eur. J.
Physiol. 463, 121–137.
Bremer, F., 1935. Cerveau “isolé” et physiologie du sommeil. C.R. Soc. Biol. 118, 1235–1241.
Bremer, F., 1936. Nouvelles recherches sur le mecanisme du sommeil. C.R. Soc. Biol. 122,
460–464.
Brown, R.E., Basheer, R., Mckenna, J.T., Strecker, R.E., Mccarley, R.W., 2012. Control of
sleep and wakefulness. Physiol. Rev. 92, 1087–1187.
References
Buonviso, N., Amat, C., Litaudon, P., Roux, S., Royet, J., Farget, V., Sicard, G., 2003. Rhythm
sequence through the olfactory bulb layers during the time window of a respiratory cycle.
Eur. J. Neurosci. 17, 1811–1819.
Carey, R., Wachowiak, M., 2011. Effect of sniffing on the temporal structure of mitral/tufted
cell output from the olfactory bulb. J. Neurosci. 31, 10615–10626.
Carmichael, S., Clugnet, M., Price, J., 1994. Central olfactory connections in the macaque
monkey. J. Comp. Neurol. 346, 403–434.
Carskadon, M.A., Herz, R.S., 2004. Minimal olfactory perception during sleep: why odor
alarms will not work for humans. Sleep 27, 402–405.
Cecchi, A., Hersant, Y., Bernard, C., 2013. La Renaissance et le rêve—Bosch, Véronèse,
Greco. RMN-Grand Palais, Paris.
Chaput, M., 1986. Respiratory-phase-related coding of olfactory information in the olfactory
bulb of awake freely-breathing rabbits. Physiol. Behav. 36, 319–324.
Cirelli, C., Tononi, G., 2008. Is sleep essential? PLoS Biol. 6, e216.
Deuker, L., Olligs, J., Fell, J., Kranz, T.A., Mormann, F., Montag, C., Reuter, M., Elger, C.E.,
Axmacher, N., 2013. Memory consolidation by replay of stimulus-specific neural activity.
J. Neurosci. 33, 19373–19383.
Diekelmann, S., Büchel, C., Born, J., Rasch, B., 2011. Labile or stable: opposing consequences
for memory when reactivated during waking and sleep. Nat. Neurosci. 14, 381–386.
Diekelmann, S., Biggel, S., Rasch, B., Born, J., 2012. Offline consolidation of memory varies
with time in slow wave sleep and can be accelerated by cuing memory reactivations. Neurobiol. Learn. Mem. 98, 103–111.
Domino, E., Ueki, S., 1959. Differential effects of general anesthetics on spontaneous electrical activity of neocortical and rhinencephalic brain systems of the dog. J. Pharmacol. Exp.
Ther. 127, 288–304.
Doty, R.L., Shaman, P., Dann, M., 1984. Development of the University of Pennsylvania smell
identification test: a standardized microencapsulated test of olfactory function. Physiol.
Behav. 32, 489–502.
Dworak, M., Mccarley, R., Kim, T., Kalinchuk, A., Basheer, R., 2010. Sleep and brain energy
levels: ATP changes during sleep. J. Neurosci. 30, 9007–9016.
Ego-Stengel, V., Wilson, M.A., 2010. Disruption of ripple-associated hippocampal activity
during rest impairs spatial learning in the rat. Hippocampus 20, 1–10.
Ellenbogen, J.M., Payne, J.D., Stickgold, R., 2006. The role of sleep in declarative memory
consolidation: passive, permissive, active or none? Curr. Opin. Neurobiol. 16, 716–722.
Faure, J., Vincent, D., 1964. The olfactory bulb in the rabbit during wakefulness and sleep.
C.R. Soc. Biol. 158, 515–519.
Field, T., Field, T., Cullen, C., Largie, S., Diego, M., Schanberg, S., Kuhn, C., 2008. Lavender
bath oil reduces stress and crying and enhances sleep in very young infants. Early Hum.
Dev. 84, 399–401.
Fontanini, A., Bower, J.M., 2005. Variable coupling between olfactory system activity and
respiration in ketamine/xylazine anesthetized rats. J. Neurophysiol. 93, 3573–3581.
Fontanini, A., Spano, P., Bower, J.M., 2003. Ketamine-xylazine-induced slow (<1.5 Hz) oscillations in the rat piriform (olfactory) cortex are functionally correlated with respiration.
J. Neurosci. 23, 7993–8001.
Franken, P., Dijk, D.-J., Tobler, I., Borb, A.A., Sleep, L.Y., 1991. Sleep deprivation spectra,
vigilance in rats: effects on EEG power states, and cortical temperature. Am. J. Physiol.
261, R198–R208.
Freud, S., 1899. Transl. A.A. Brill. The Interpretation of Dreams. The Macmillan Company,
New York, 1913.
339
340
CHAPTER 12 Smell, Sleep, and Memory
Friedman, L., Bergmann, B.M., Rechtschaffen, A., 1979. Effects of sleep deprivation on sleepiness, sleep intensity, and subsequent sleep in the rat. Sleep 1, 369–391.
Fuentemilla, L., Miró, J., Ripollés, P., Vilà-Balló, A., Juncadella, M., Castañer, S., Salord, N.,
Monasterio, C., Falip, M., Rodrı́guez-Fornells, A., 2013. Hippocampus-dependent
strengthening of targeted memories via reactivation during sleep in humans. Curr. Biol.
CB 23, 1769–1775.
Fuller, P.M., Gooley, J.J., Saper, C.B., 2006. Neurobiology of the sleep-wake cycle: sleep architecture, circadian regulation, and regulatory feedback. J. Biol. Rhythm. 21, 482–493.
Gaiman, N., 1991–1996. The Sandman, vols. 1–10. Vertigo/DC Comics, New York City,
New York.
Gervais, R., Pager, J., 1979. Combined modulating effects of the general arousal and the specific hunger arousal on the olfactory bulb responses in the rat. Electroencephalogr. Clin.
Neurophysiol. 46, 87–94.
Gervais, R., Pager, J., 1982. Functional changes in waking and sleeping rats after lesions in the
olfactory pathways. Physiol. Behav. 29, 7–15.
Girardeau, G., Benchenane, K., Wiener, S.I., Buzsáki, G., Zugaro, M.B., 2009. Selective suppression of hippocampal ripples impairs spatial memory. Nat. Neurosci. 12, 1222–1223.
Goel, N., Lao, R.P., 2006. Sleep changes vary by odor perception in young adults. Biol. Psychol. 71, 341–349.
Goel, N., Kim, H., Lao, R.P., 2005. An olfactory stimulus modifies nighttime sleep in young
men and women. Chronobiol. Int. 22, 889–904.
Gottfried, J.A., 2006. Smell: central nervous processing. Adv. Otorhinolaryngol. 63, 44–69.
Gottfried, J.A., 2010. Central mechanisms of odour object perception. Nat. Rev. Neurosci. 11,
628–641.
Grosmaitre, X., Santarelli, L.C., Tan, J., Luo, M., Ma, M., 2007. Dual functions of mammalian
olfactory sensory neurons as odor detectors and mechanical sensors. Nat. Neurosci. 10,
348–354.
Grosmaitre, X., Fuss, S., Lee, A., Adipietro, K., Matsunami, H., Mombaerts, P., Ma, M., 2009.
SR1, a mouse odorant receptor with an unusually broad response profile. J. Neurosci. 29,
14545–14552.
Grupp, K., Maurer, J.T., Hörmann, K., Hummel, T., Stuck, B.A., 2008. Chemosensory induced
arousals during sleep in premenopausal women. Neurosci. Lett. 444, 22–26.
Hauner, K.K., Howard, J.D., Zelano, C., Gottfried, J.A., 2013. Stimulus-specific enhancement
of fear extinction during slow-wave sleep. Nat. Neurosci. 16, 1553–1555.
Hernández-Peón, R., Lavin, A., Alcocer-Cuarón, C., Marcelin, J., 1960. Electrical activity of
the olfactory bulb during wakefulness and sleep. EEG Clin. Neurophysiol. 12, 41–58.
Herry, C., Ciocchi, S., Senn, V., Demmou, L., Muller, C., Luthi, A., 2008. Switching on and off
fear by distinct neuronal circuits. Nature 454, 600–606.
Iber, C., Ancoli-Israel, S., Chesson, A., Quan, S., 2007. The AASM Manual for the Scoring
of Sleep and Associated Events: Rules, Terminology and Technical Specifications.
American Academy of Sleep Medicine, Westchester.
Inostroza, M., Born, J., 2013. Sleep for preserving and transforming episodic memory. Annu.
Rev. Neurosci. 36, 79–102.
Jenkins, J., Dallenbach, K., 1924. Obliviscence during sleep and waking. Am. J. Psychol. 35,
605–612.
Johnson, B., Mainland, J., Sobel, N., 2003. Rapid olfactory processing implicates subcortical
control of an olfactomotor system. J. Neurophysiol. 90, 1084–1094.
References
Kavanau, J.L., 2006. Is sleep’s ‘supreme mystery’ unraveling? An evolutionary analysis of
sleep encounters no mystery; nor does life’s earliest sleep, recently discovered in jellyfish.
Med. Hypotheses 66, 3–9.
Keville, K., Green, M., 1995. Aromatherapy: A Complete Guide to the Healing Art. The
Crossing Press, Freedom, CA.
Killgore, W.D.S., Mcbride, S.A., 2006. Odor identification accuracy declines following 24 h
of sleep deprivation. J. Sleep Res. 15, 111–116.
Knutson, K.L., Spiegel, K., Penev, P., Van Cauter, E., 2007. The metabolic consequences of
sleep deprivation. Sleep Med. Rev. 11, 163–178.
Lange, T., Dimitrov, S., Born, J., 2010. Effects of sleep and circadian rhythm on the human
immune system. Ann. N. Y. Acad. Sci. 1193, 48–59.
Leshan, L., 1942. The breaking of a habit by suggestion during sleep. J. Abnorm. Psychol. 37,
406–408.
Lewith, G., Godfrey, A., Prescott, P., 2005. A Single-Blinded, Randomized Pilot Study Evaluating the Aroma of Lavandule augustifolia as a Treatment for Mild Insomnia. J. Altern.
Complement. Med. 11, 631–637.
Lyamin, O.I., Mukhametov, L.M., Siegel, J.M., Nazarenko, E.A., Polyakova, I.G.,
Shpak, O.V., 2002. Unihemispheric slow wave sleep and the state of the eyes in a white
whale. Behav. Brain Res. 129, 125–129.
Manabe, H., Kusumoto-Yoshida, I., Ota, M., Mori, K., 2011. Olfactory cortex generates synchronized top-down inputs to the olfactory bulb during slow-wave sleep. J. Neurosci. 31,
8123–8133.
Mcbride, S.A., Balkin, T.J., Kamimori, G.H., Killgore, W.D.S., 2006. Olfactory decrements as
a function of two nights of sleep deprivation. J. Sens. Stud. 21, 456–463.
Morselli, L.L., Guyon, A., Spiegel, K., 2012. Sleep and metabolic function. Pflüg. Arch. Eur. J.
Physiol. 463, 139–160.
Mukhametov, L.M., Supin, A.Y., Polyakova, I.G., 1977. Interhemispheric asymmetry of the
electroencephalographic sleep patterns in dolphins. Brain Res. 134, 581–584.
Murakami, M., Kashiwadani, H., Kirino, Y., Mori, K., 2005. State-dependent sensory gating in
olfactory cortex. Neuron 46, 285–296.
Murthy, V.N., 2011. Olfactory maps in the brain. Annu. Rev. Neurosci. 34, 233–258.
Oudiette, D., Paller, K.A., 2013. Upgrading the sleeping brain with targeted memory reactivation. Trends Cogn. Sci. 17, 142–149.
Oudiette, D., Antony, J.W., Creery, J.D., Paller, K.A., 2013a. The role of memory reactivation
during wakefulness and sleep in determining which memories endure. J. Neurosci. 33,
6672–6678.
Oudiette, D., Antony, J.W., Paller, K.A., 2013b. Fear not: manipulating sleep might help you
forget. Trends Cogn. Sci. 18, 3–4.
Pavlides, C., Winson, J., 1989. Influences of hippocampal place cell firing in the awake
state on the activity of these cells during subsequent sleep episodes. J. Neurosci. 9,
2907–2918.
Pennacchio, M., Jefferson, L., Havens, K., 2010. Uses and Abuses of Plant-Derived Smoke: Its
Ethnobotany as Hallucinogen, Perfume, Incense, and Medicine. Oxford University Press,
New York.
Rasch, B., Born, J., 2013. About sleep’s role in memory. Physiol. Rev. 93, 681–766.
Rasch, B., Büchel, C., Gais, S., Born, J., 2007. Odor cues during slow-wave sleep prompt declarative memory consolidation. Science (N.Y.) 315, 1426–1429.
341
342
CHAPTER 12 Smell, Sleep, and Memory
Ray, J.P., Price, J.L., 1992. The organization of the thalamocortical connections of the mediodorsal thalamic nucleus in the rat, related to the ventral forebrain–prefrontal cortex topography. J. Comp. Neurol. 323, 167–197.
Rechtschaffen, A., Bergmann, B.M., 2002. Sleep deprivation in the rat: an update of the 1989
paper. Sleep 25, 18–24.
Rechtschaffen, A., Kales, A., 1968. A Manual of Standardized Terminology, Techniques and
Scoring System for Sleep Stages of Human Subjects. U.S. Dept. of Health, Education, and
Welfare, Public Health Service, Bethesda, MD.
Rolls, A., Makam, M., Kroeger, D., Colas, D., De Lecea, L., Heller, H.C., 2013. Sleep to forget: interference of fear memories during sleep. Mol. Psychiatry 18, 1166–1170.
Rudoy, J.D., Voss, J.L., Westerberg, C.E., Paller, K.A., 2009. Strengthening individual memories by reactivating them during sleep. Science (N.Y.) 326, 1079.
Russchen, F.T., Amaral, D.G., Price, J.L., 1987. The afferent input to the magnocellular division of the mediodorsal thalamic nucleus in the monkey, Macaca fascicularis. J. Comp.
Neurol. 256, 175–210.
Seymour, J.E., Carrette, T.J., Sutherland, P.A., 2004. Do box jellyfish sleep at night? Med. J.
Aust. 181, 707.
Shusterman, R., Smear, M., Koulakov, A., Rinberg, D., 2011. Precise olfactory responses tile
the sniff cycle. Nat. Neurosci. 14, 1039–1044.
Siegel, J., 2002. The Neural Control of Sleep and Waking. Springer-Verlag, New York.
Stephenson, R., Chu, K.M., Lee, J., 2007. Prolonged deprivation of sleep-like rest raises metabolic rate in the Pacific beetle cockroach, Diploptera punctata (Eschscholtz). J. Exp. Biol.
210, 2540–2547.
Stuck, B.A., Weitz, H., Hörmann, K., Maurer, J.T., Hummel, T., 2006. Chemosensory eventrelated potentials during sleep—a pilot study. Neurosci. Lett. 406, 222–226.
Stuck, B.A., Stieber, K., Frey, S., Freiburg, C., Hörmann, K., Maurer, J.T., Hummel, T., 2007.
Arousal responses to olfactory or trigeminal stimulation during sleep. Sleep 30, 506–510.
Tambini, A., Davachi, L., 2013. Persistence of hippocampal multivoxel patterns into postencoding rest is related to memory. Proc. Natl. Acad. Sci. U. S. A. 110, 19591–19596.
Tambini, A., Ketz, N., Davachi, L., 2010. Enhanced brain correlations during rest are related to
memory for recent experiences. Neuron 65, 280–290.
Tanabe, T., Yarita, H., Lino, M., Ooshima, Y., Takagi, S.F., 1975. An olfactory projection
area in orbitofrontal cortex of the monkey. APJ J. Physiol. 38, 1269–1283.
Thomas, M., Sing, H., Belenky, G., Holcomb, H., Mayberg, H., Dannals, R., Wagner, H.,
Thorne, D., Popp, K., Redmond, D., 2000. Neural basis of alertness and cognitive performance impairments during sleepiness. I. Effects of 24 h of sleep deprivation on waking
human regional brain activity. J. Sleep Res. 9, 335–352.
Tononi, G., Cirelli, C., 2003. Sleep and synaptic homeostasis: a hypothesis. Brain Res. Bull.
62, 143–150.
Tononi, G., Cirelli, C., 2006. Sleep function and synaptic homeostasis. Sleep Med. Rev. 10,
49–62.
Van Cauter, E., Spiegel, K., Tasali, E., Leproult, R., 2008. Metabolic consequences of sleep
and sleep loss. Sleep Med. 9 (Suppl. 1), S23–S28.
Van Dongen, H.P.A., Maislin, G., Mullington, J.M., Dinges, D.F., 2003. The cumulative
cost of additional wakefulness: dose–response effects on neurobehavioral functions and
sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 26,
117–126.
References
Van Dongen, E.V., Takashima, A., Barth, M., Zapp, J., Schad, L.R., Paller, K.A.,
Fernández, G., 2012. Memory stabilization with targeted reactivation during human
slow-wave sleep. Proc. Natl. Acad. Sci. U. S. A. 109, 10575–10580.
Velluti, R.A., 1997. Interactions between sleep and sensory physiology. J. Sleep Res. 6, 61–77.
Wachowiak, M., 2011. All in a sniff: olfaction as a model for active sensing. Neuron 71,
962–973.
Walker, M.P., Stickgold, R., 2006. Sleep, memory, and plasticity. Annu. Rev. Psychol. 57,
139–166.
Wang, S., Teixeira, C., Wheeler, A., Frankland, P., 2009. The precision of remote context
memories does not require the hippocampus. Nat. Neurosci. 12, 253–255.
Wang, G., Grone, B., Colas, D., Appelbaum, L., Mourrain, P., 2011. Synaptic plasticity in
sleep: learning, homeostasis and disease. Trends Neurosci. 34, 452–463.
Wilson, D.A., 2010. Single-unit activity in piriform cortex during slow-wave state is shaped by
recent odor experience. J. Neurosci. 30, 1760–1765.
Wilson, M., Mcnaughton, B., 1994. Reactivation of hippocampal ensemble memories during
sleep. Science 265, 5–8.
Wilson, D.A., Sullivan, R.M., 2011. Cortical processing of odor objects. Neuron 72, 506–519.
Wilson, D.A., Yan, X., 2010. Sleep-like states modulate functional connectivity in the rat olfactory system. J. Neurophysiol. 104, 3231–3239.
Wilson, D.A., Hoptman, M.J., Gerum, S.V., Guilfoyle, D.N., 2011. State-dependent functional
connectivity of rat olfactory system assessed by fMRI. Neurosci. Lett. 497, 69–73.
Yarita, H., Lino, M., Tanabe, T., Kogure, S., Takagi, S., 1980. A transthalamic olfactory pathway to orbitofrontal cortex in the monkey. APJ J. Physiol. 43, 69–85.
Yokoyama, T.K., Mochimaru, D., Murata, K., Manabe, H., Kobayakawa, K., Kobayakawa, R.,
Sakano, H., Mori, K., Yamaguchi, M., 2011. Elimination of adult-born neurons in the olfactory bulb is promoted during the postprandial period. Neuron 71, 883–897.
343