JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY, 2017

VOL. 39, NO. 7, 646–658

[Link]

Association between olfaction and higher cortical functions in Alzheimer’s

disease, mild cognitive impairment, and healthy older adults
Amanda M. Warda, Matthew Calamiab, Erin Thiemanna, Jamie Dunlapa and Daniel Tranela,c
a
Department of Psychological and Brain Sciences, University of Iowa, Iowa City, IA, USA; bDepartment of Psychology, Louisiana State
University, Baton Rouge, LA, USA; cDepartment of Neurology, University of Iowa Carver College of Medicine, Iowa City, IA, USA

ABSTRACT ARTICLE HISTORY

Introduction: Neural regions important for smell are proximal and closely connected to cortical Received 4 April 2016
areas that have been strongly implicated in higher order functions of value-based decision Accepted 17 October 2016
making and emotional memory. The integrity of these neural regions are affected in aging and KEYWORDS
neurodegenerative conditions. Two specific predictions follow from these neuroanatomical Decision making; Dementia;
arrangements—namely, that olfaction would be associated with value-based decision making Emotions; Memory; Smell
and with emotional memory. Method: To test these predictions, we measured these different
capacities in participants with presumed varying degrees of integrity of the relevant brain
structures: specifically, 13 patients with Alzheimer’s disease, 8 patients with mild cognitive
impairment, and 20 healthy older adults. The participants completed detailed tests of olfaction,
value-based decision making, emotional memory, and general cognitive ability. Results: Olfactory
functioning was significantly associated with emotional and nonemotional memory. The associa-
tion was especially strong and consistent for memory recall with olfaction, explaining as much as
10% additional variance over and above general cognition. Olfactory functioning was not
strongly or consistently associated with decision making over and above general cognition.
Conclusion: Olfaction is a strong predictor of memory recall. These findings may contribute to a
better understanding of olfaction and specific cognitive domains known to be affected by aging
and implicated in neurodegenerative disease.

Olfactory functioning has been found to decline in et al., 2000). Despite the potential clinical utility of
healthy older adults (Makowska, Kloszewska, measuring olfaction, the relationship between olfaction
Grabowska, Szatkowska, & Rymarczyk, 2011; Murphy and cognition remains incompletely understood.
et al., 2002) and to be impaired in various populations of While impaired olfactory performance is related to a
patients with neurodegenerative conditions. Previous diagnosis of MCI and AD, the relationships between
research has found odor identification (Doty, Reyes, & olfaction and specific domains of cognitive functioning
Gregor, 1987; Kesslak et al., 1988; Koss, 1986; Makowska have been less explored. In general, smell identification
et al., 2011; Warner, Peabody, Flattery, & Tinklenberg, has been shown to be related to performance on var-
1986), odor memory (Kesslak et al., 1988; Naudin et al., ious neuropsychological tests, such as tests measuring
2014), and odor threshold deficits (Djordjevic, Jones- general cognition, memory, language, and executive
Gotman, De Sousa, & Chertkow, 2008; Murphy, functioning, in healthy older adults (Royall, Chiodo,
Gilmore, Seery, Salmon, & Lasker, 1990; Nordin, Polk, & Jaramillo, 2002), patients with MCI
Monsch, & Murphy, 1995) in patients with (Devanand et al., 2010; Djordjevic et al., 2008; Fusetti
Alzheimer’s disease (AD). Olfaction was also impaired et al., 2010; Lehrner, Pusswald, Gleiss, Auff, & Dal-
in patients with mild cognitive impairment (MCI) in Bianco, 2009; Lojkowska et al., 2011; Makizako et al.,
terms of odor identification (Bahar-Fuchs, Moss, Rowe, 2014; Vyhnalek et al., 2015), and patients with AD
& Savage, 2011; Devanand et al., 2010; Djordjevic et al., (Djordjevic et al., 2008; Makowska et al., 2011).
2008; Eibenstein et al., 2005) and odor threshold Previous research has investigated the relationship
(Djordjevic et al., 2008). Olfaction predicts the transition between olfaction and autobiographical memory in
to MCI (Wilson et al., 2007) and the transition to AD healthy individuals and found that olfactory cues
(Conti et al., 2013; Devanand et al., 2015; Devanand evoke more detailed (Chu & Downes, 2002) and

CONTACT Amanda M. Ward Amanda-m-ward@[Link] Department of Psychological and Brain Sciences, University of Iowa, E11 Seashore Hall,
Iowa City, IA 52242, USA
© 2016 Informa UK Limited, trading as Taylor & Francis Group
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 647

emotional autobiographical memories (Herz, 2004; older adults, patients with MCI, and patients with AD.
Herz, Eliassen, Beland, & Souza, 2004; Willander & This approach captures a range of olfactory and cog-
Larsson, 2007). However, there are two key areas of nitive abilities and (presumably) varying degrees of
cognition that are especially relevant in AD and in MCI neural integrity.
that have not been explored.
Aspects of higher functioning that are especially
intriguing in this context are emotional memory and Method
decision making. Emotional memory includes retrieval
Participants
of previously experienced or presented emotionally
valenced stimuli (Buchanan, 2007). Value-based deci- Thirteen patients with AD, 8 patients with MCI, and 20
sion making involves the development of emotional healthy older adults participated in the current study.
biases from previous experiences that are used to Fourteen of the AD and MCI patients were recruited
make effective decisions (Bechara, Damasio, Tranel, & through the Benton Neuropsychology Clinic at
Damasio, 1997). Emotional memory and decision mak- University of Iowa. Clinical neuropsychological reports
ing involve neural processing in the anterior medial indicated that these patients were diagnosed with mild
temporal lobe (Buchanan, 2007; Dolcos, LaBar, & to moderate AD (n = 7) and MCI (n = 7) according to
Cabeza, 2004) and the orbitofrontal cortex, respectively established criteria (Albert et al., 2011; McKhann et al.,
(Bechara, Damasio, & Damasio, 2000; Strenziok et al., 2011) in the Department of Neurology at the
2011), which are regions also important for olfaction University of Iowa. The remainder of individuals with
(Buchanan, Tranel, & Adolphs, 2003; Dade, Jones- cognitive decline (n = 6 patients with AD and n = 1
Gotman, Zatorre, & Evans, 1998; Savic, 2002; Wang, patient with MCI) were recruited through the
Eslinger, Smith, & Yang, 2005). There is evidence that Alzheimer’s Association. These patients were diag-
suggests that both emotional memory (Fujie et al., nosed by a neurologist or neuropsychologist at another
2008; Hamann, Monarch, & Goldstein, 2000; institution. Healthy older adults were recruited through
Kensinger, Brierley, Medford, Growdon, & Corkin, the Cognitive Neuroscience Registry for Normative
2002) and decision making (Delazer, Sinz, Zamarian, Data at the University of Iowa.
& Benke, 2007; Sinz, Zamarian, Benke, Wenning, & For all participants (patients and healthy), exclusion
Delazer, 2008; Zamarian, Weiss, & Delazer, 2011) are criteria included neurological disease besides AD or
impaired in MCI and AD. Due to an early decline in MCI, intellectual disability, history of a learning dis-
both olfactory functioning and both emotional mem- ability, severe depression or other psychiatric disease,
ory and decision making in patients with MCI and AD, uncorrected impaired vision or hearing, impaired com-
it is possible that a strong association might be prehension, impaired basic attention, and nasal condi-
expected between olfaction and these domains of cog- tions that interfere with the ability to smell. In healthy
nitive functioning. The aims of the current study were older adults, exclusion criteria also included a diagnosis
(a) to investigate the relationship between olfactory of MCI or AD. Eligibility of patients seen in the Benton
functioning and value-based decision making and emo- Neuropsychology Clinic was determined through
tional memory over and above general cognition, and patient and caregiver interviews via the telephone,
(b) to compare the strength of the association between review of medical record, and performance on neurop-
olfaction and emotional memory and olfaction and sychological measures. Eligibility of patients recruited
nonemotional memory. It was hypothesized that olfac- from other institutions was determined through patient
tory functioning would be significantly associated with and/or caregiver interviews via the telephone. In each
value-based decision making and emotional memory phone call, the patient and/or caregiver were asked
over and above general cognition. Secondly, it was whether patients had a diagnosis of AD or MCI by a
hypothesized that olfaction would have a stronger asso- neurologist or a neuropsychologist. They were also
ciation with emotional memory than with nonemo- screened for the remaining exclusion criteria. While
tional memory. In order to test the associations we did not have any other background or medical
between olfaction and higher cognitive functioning, information on these patients besides what was col-
variability was needed in each of the domains. The lected on the demographics form, the patients with
range of scores in olfaction and higher cognitive func- AD (6) and MCI (1) had Montreal Cognitive
tioning would be restricted if it was examined in each Assessment (MOCA) scores in the impaired range
group individually; thus the associations between olfac- (mean = 11.14, range = 5–18).
tion and each higher cognitive functioning variable Regarding medications, two of the healthy partici-
were examined across all the groups, including healthy pants were taking an antidepressant medication. In
648 A. M. WARD ET AL.

patients with MCI, three patients were on a central Mood

acetylcholinesterase inhibitor, five were on antidepres- The Beck Depression Inventory Second Edition is a 21-
sant medication, and one was on a benzodiazepine item self-report measure of depressive symptoms that
medication. In patients with AD, two patients were was administered to every participant (Beck, Steer, &
on both a central acetylcholinesterase inhibitor and Brown, 1996). Higher scores are indicative of higher
an NMDA receptor antagonist, one patient was on an depressive symptomatology.
antidepressant medication, and two patients were on
benzodiazepine medication. Olfaction
All participants gave written consent to the protocol Smell functioning was assessed using two smell tests:
approved by the University of Iowa Institutional The University of Pennsylvania Smell Identification
Review Board. The study was conducted in compliance Test (UPSIT) and the Odor Memory Test. The UPSIT
with the Helsinki Declaration of 1975. A standardized is a 40-item four-alternative forced-choice test with
procedure (DeRenzo, Conley, & Love, 1998) was used microencapsulated odorants in scratch and sniff format
to determine whether the patients were able to consent. (Doty, 1983). It has a high test–retest reliability of .90
If patients were unable to consent, caregivers provided (Doty, McKeown, Lee, & Shaman, 1995). The Odor
informed consent while patients signed an assent Identification Task was scored based on the raw score
document. correctly identified.
The Odor Memory Test is a 12-item four-alternative
Demographic information forced-choice test with microencapsulated odorants in
Demographic information was collected through a scratch and sniff format (Doty, 2003). The test includes
questionnaire. Information was collected about the the presentation of an odorant and a delay interval, and
patient’s age, education, gender, and smoking history. then participants are asked to select the odor that was
Smoking was assessed using pack-years, which is cal- previously presented. The test has an adequate test–
culated by multiplying the packs of cigarettes smoked retest reliability of .68 (Doty et al., 1995). The Odor
per day by the number of years the individual has Memory Task was scored based on the total raw score
smoked. These variables were collected to better char- correctly identified. Even though the Odor Memory
acterize the current sample. Test involves three different types of delay intervals
(10, 30, or 60 seconds), we used the total raw score
Cognition and reading ability since previous research has used the total score (Jiang
The MOCA was administered as a measure of general et al., 2008; Tourbier & Doty, 2007), and previous
cognitive ability (Nasreddine et al., 2005). Scores range research has not found a change in performance across
from 0 to 30 with higher scores indicating better gen- delay intervals (Choudhury, Moberg, & Doty, 2003;
eral cognitive ability. The Auditory Verbal Learning Doty, Kisat, & Tourbier, 2008). Also, the test–retest
Test (AVLT) Short Delayed Recall (Rey, 1958) was reliability has been completed only on the total score
given in a clinical context in the Benton (Doty et al., 1995).
Neuropsychology Clinic at University of Iowa. Since In order to standardize and ease administration, the
this measure was not administered in a research con- examiner verbally read each item aloud and scratched
text, the measure was not given to all participants. the label for each participant for both the UPSIT and
Thus, 11 patients with Alzheimer’s disease and 1 the Odor Memory Test. Four items overlap between
patient with MCI had data on AVLT Short Delayed the UPSIT and the Odor Memory Test. In our proce-
Recall using the raw score. AVLT Short Delayed Recall dure, there was at least 30 minutes between each test in
was chosen since previous research assessing the rela- order to prevent interference between the two tasks.
tionship between olfaction and memory commonly The UPSIT contains 40 items, so the likely impact of
used delayed memory recall measures. the overlapping items would be minimal. Thus, we did
The Wechsler Test of Adult Reading (WTAR) is a not make any corrections for this overlap.
word reading test used to estimate premorbid function-
ing (Wechsler, 2001); however, performance on this Decision making
measure declines with AD (McFarlane, Welch, & The decision-making tasks conducted included the
Rodgers, 2006). As olfaction is not expected to relate Iowa Gambling Task and the Balloon Analogue Risk
to word reading, the WTAR was included to demon- Task. Both tasks simulate real-world decisions through
strate the potential specificity of olfactory impairment uncertainty, rewards, and punishment. The Iowa
to emotional memory and decision making. The Gambling Task (Bechara et al., 1997) is a computer-
WTAR was scored using age-corrected standard scores. based decision-making task, which involves the
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 649

selection of cards from four decks. The selection of a significant difference between patients with AD, those
card results in a monetary reward or a monetary with MCI, and healthy older adults in ratings of
reward and a penalty. Participants are told that some valence or arousal (p > .05). Then pictures were pre-
decks are more advantageous than others, and the sented a second time without any ratings. Immediately
point of the game is to win as much money as possible. after the two presentations of pictures, participants
Selecting the advantageous decks results in a net gain, were given a free recall test and a cued recall test.
while selecting the disadvantageous decks results in a Participants were not aware of the memory tests
net loss. The Iowa Gambling Task score was calculated beforehand. During the free recall task, participants
in two ways (Sinz et al., 2008; Zamarian et al., 2011). were told to provide a short description of as many
First, a net score for each participant was calculated by pictures as they could. After the free recall task, parti-
subtracting the number of selected bad decks from the cipants were provided with cue words to allow for
number of selected good decks. The score can range further recall. A cued recall portion was included to
from −100 to 100. Secondly, shifts between decks was prevent floor effects in the performance of patients
calculated. In previous research, fewer shifts between with AD. The cued recall task was based on a pre-
decks reflected better strategy development (Sinz et al., viously developed cued recall task (Buchanan &
2008). The score can range from 0 to 99. Lovallo, 2001) and included the following descriptors:
The Balloon Analogue Risk Task (BART; Lejuez injured people, animals, household objects, transporta-
et al., 2002) is a computerized risk assessment task. tion, celebration, structures/architecture, and school.
Participants are told that they can win money while All the pictures could be categorized into these seven
pumping up a balloon on a computer screen and that descriptors. Pictures were counted as correct if the
they earn money with each pump. If they push the stop description applied exclusively to one picture in both
button before the balloon pops, they are able to win the the free and cued recall task. Positive, negative, and
money on that trial. However, if the balloon pops, they neutral recall scores were calculated for each partici-
are not rewarded the money on that trial. Thus, the pant by calculating the number of positive, negative,
appropriate amount of risk-taking behavior allows and neutral pictures that were recalled in total for both
access to more of the reward. There are 30 trials in the free and cued recall task, with 20 being the max-
total. The Balloon Analogue Risk Task was scored imum recalled in each category.
using the adjusted number of pumps, which is the During the recognition task, participants were pre-
average number of pumps without exploded balloons. sented with 120 pictures. Sixty pictures were presented
This method is what is most commonly used in the previously, and 60 pictures were foils. The foils were
literature since the number of pumps is constrained by mirror images of the original photos. An original pic-
the exploded balloons, and thus the adjusted number of ture and its foil were presented at once, and partici-
pumps is a more accurate representation of the amount pants were instructed to choose the picture that was
of pumps completed by participants (Lejuez et al., presented previously. Positive, negative, and neutral
2002). recognition scores were calculated for each participant
by calculating the number of positive, negative, and
Emotional and neutral memory neutral pictures that were recognized, with 20 being
The emotional memory materials and task were based the maximum recognized in each category.
on previously established emotional memory tasks
(Hamann et al., 2000; Kensinger et al., 2002). The
Procedure
emotional memory stimuli were chosen from the
International Affective Picture System (IAPS; Lang, Participants completed the research tasks in a session
Bradley, & Cuthbert, 2008). Sixty pictures were chosen that lasted approximately 4 hours. They began with the
based on the IAPS ratings of valence and arousal and BDI. Then there were two presentations of pictures,
were placed into three categories: 20 with relatively and a cued and free recall task. After the conclusion of
positive valence and high arousal, 20 with relatively the recall tasks, a break occurred where participants
negative valence and high arousal, and 20 with rela- completed the MOCA and WTAR. Afterwards, the
tively neutral valence and low arousal. emotion recognition task occurred. Then participants
The pictures were used in recall and recognition completed the Odor Memory Test, the BART, the Iowa
tasks. The sixty pictures were presented twice to parti- Gambling Task (IGT), and the UPSIT. If requested by
cipants for 6 seconds to avoid floor effects. The first the participant, the research session was divided into
time pictures were presented, participants rated each two separate testing sessions completed no longer than
picture on a scale of valence and arousal. There was no a week apart.
650 A. M. WARD ET AL.

Data analyses variable over and above general cognitive ability (as
measured by MOCA). These analyses were completed
Statistical analyses were completed using SPSS Version
on the entire sample. Each olfactory variable was exam-
19 (SPSS Inc., Chicago, IL). A Kruskal–Wallis H test or
ined in separate regression models given the high cor-
χ2, as appropriate, was conducted to determine whether
relation between the UPSIT and the Odor Memory
there were differences in demographic variables (age,
Test (r = .67).
education, gender, pack-years, depression, MOCA, and
A Fisher r-to-z transformation was conducted to
reading ability), olfaction (odor identification, odor
compare the strength of the relationship between olfac-
memory), memory (positive recall, negative recall, neu-
tion (odor identification, odor memory) and emotional
tral recall, positive recognition, negative recognition,
memory (positive recall, negative recall, positive recog-
neutral recognition), and decision making (IGT perfor-
nition, negative recognition) with the strength of the
mance and BART performance) between healthy older
relationship between olfaction (odor identification,
adults, MCI, and AD. Holm’s sequential Bonferroni
odor memory) and neutral memory (neutral recall,
procedure was used to correct for multiple compari-
neutral recognition). Thus, a total of eight comparisons
sons when performing pairwise comparisons after sig-
were made. The Pearson r values were used to com-
nificant results. Effect sizes were calculated using
plete these comparisons.
Cohen’s d (Cohen, 1988). Bivariate Spearman correla-
tions were conducted to predict a clinical measure of
memory (AVLT Short Delayed Recall) to confirm pre-
vious research examining the association between Results
olfaction and memory recall, but was not explored Table 1 includes demographic and neuropsychological
further due to a small sample size. data. Regarding demographic data, there were no sig-
Bivariate Pearson correlations were conducted to nificant differences between groups on age, education,
predict the following higher cognitive functioning vari- gender, pack-years, depression, or reading ability
ables from olfaction: positive, negative, and neutral (p > .05). Participants experienced minimal to moderate
recall; positive, negative, and neutral recognition; symptoms of depression, as assessed by the Beck
BART performance; IGT performance; and reading Depression Inventory (Beck, Ward, Mendelson, Mock,
performance. These analyses were completed on the & Erbaugh, 1961). As expected, MOCA scores were
entire sample including healthy older adults, MCI, significantly different between groups (p < .005).
and AD in order to analyze a wide spectrum of olfac- Pairwise comparisons were performed using Holm’s
tory and cognitive abilities. If the association was sig- sequential Bonferroni procedure to correct for multiple
nificant, the association was explored further in a comparisons. This post hoc analysis revealed statistically
regression model. significant differences in MOCA scores between AD and
Hierarchical multiple regression models were run on healthy older adults (p < .001, d = 2.34) and between
the significant bivariate associations to predict the MCI and healthy older adults (p < .005, d = 2.28) but no
higher cognitive functioning variables from olfaction. significant difference between the AD and MCI
Hierarchical multiple regression models were run to (p > .05). For the patients that were given the AVLT,
determine whether the addition of olfaction improved they had a performance with a mean of 3.5 words
the prediction of each higher cognitive functioning remembered (SD = 1.6).

Table 1. Demographic characteristics of healthy elderly, MCI participants, and AD participants.

Total AD MCI HE
participants n = 13 M n=8M n = 20 M Overall
Characteristics (n = 41) (SD) (SD) (SD) Kruskal–Wallis p Pairwise comparison p and effect sizes
Age (years) 76.59 (6.44) 76.77 (6.95) 76.13 (6.29) 76.65 (6.48) .957 —
Education (years) 16.40 (3.18) 16.08 (3.59) 15.5 (3.51) 16.98 (2.80) .448 —
No. of females 21 9 3 9 .272 —
Pack-years 6.85 (14.95) 3.41(10.19) 4.92 (7.27) 9.86 (19.14) .301 —
BDI 4.34 (4.97) 4.31 (3.17) 8.75 (8.81) 2.60 (2.37) .132 —
MOCA 20.66 (7.19) 13.92 (6.81) 18.00 (4.17) 26.10 (2.81) <.001 AD vs. HE p < .001, d = 2.34
MCI vs. HE, p < .01, d = 2.28
WTAR 107.80 (13.46) 100.77 (17.81) 107.25 (10.15) 112.60 (9.21) .140 —
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment; HE = healthy elderly; BDI = Beck Depression Inventory;
MOCA = Montreal Cognitive Assessment; WTAR = Wechsler Test of Adult Reading. Data are presented as means (standard deviations in parentheses)
(Cohen, 1988). Effect sizes were calculated using Cohen’s d. The effect sizes follow the 0.2 small, 0.5 medium, and 0.8 strong effect size.
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 651

Differences between healthy older adults, MCI, and and MCI participants (p > .05), or between MCI and
AD in olfaction, memory, and decision making healthy older adults (p > .05). There was no difference
between groups in terms of Iowa Gambling Task per-
Mean values for memory and olfaction performance
formance using net score (p > .05), but healthy older
for each group (healthy older adults, MCI, and AD) are
adults had lower shifts between decks when compared
presented in Table 2. There were significant differences
to AD (p < .05, d = −1.21). There was no difference
between groups regarding odor identification and odor
between groups in terms of Balloon Analog Risk Task
memory. Odor identification performance was higher
performance (p > .05).
in healthy older adults when compared to both MCI
We examined the association between each olfactory
(p < .05, d = 1.58) and AD participants (p < .05,
variable and AVLT using a Spearman correlation and
d = 3.26). Odor memory performance was higher in
found that both odor identification (r = .61, p < .05)
healthy older adults when compared to AD partici-
and odor memory (r = .59, p < .05) were significantly
pants (p < .05, d = 1.28), but not when compared to
related to AVLT Short Delayed Recall.
MCI participants (p > .05). There were no significant
differences between MCI and AD in terms of both
odor identification (p > .05) and odor mem-
Relationship between olfaction and emotional
ory (p > .05).
memory
There were significant differences between groups in
terms of recall and recognition. Positive recall perfor- Pearson correlations were completed between each
mance was higher in healthy older adults when com- olfactory variable and each higher cognitive function-
pared to MCI (p < .05, d = 1.79) and AD participants ing variable and are presented in Table 3. Odor identi-
(p < .05, d = 2.64). No difference was found between fication and odor memory were significantly related to
MCI and AD (p > .05). Performance was higher in positive recall, negative recall, positive recognition, and
healthy older adults when compared to AD on negative negative recognition (p < .05).
recall (d = 3.80), neutral recall (d = 2.40), positive Hierarchical regression models were completed for
recognition (d = 1.77), negative recognition each olfactory variable as the independent variables
(d = 2.58), and neutral recognition (d = 2.14), and positive recall as the outcome variable. General
p < .05, but there were no differences between AD cognition was entered into the first block of each

Table 2. Olfaction, memory, and decision-making performance.

Total participants AD MCI HE
n = 41 n = 13 M n=8 n = 20 M
Variables M (SD) (SD) M (SD) (SD) Overall p Pairwise comparison p and effect sizes
UPSIT 25.90 (9.86) 16.69 (6.51) 21.63 (10.17) 33.60 (3.39) <.001 AD vs. HE p < .001, d = 3.26
MCI vs. HE, p < .05, d = 1.58
OMT 5.66 (3.13) 3.92 (1.85) 4.38 (2.45) 7.30 (3.26) <.01 AD vs. HE, p < .01, d = 1.28
Positive recall 6.83 (5.18) 2.31 (2.14) 4.25 (3.28) 10.80 (4.02) <.001 AD vs. HE, p < .001, d = 2.64
MCI vs. HE, p < .02, d = 1.79
Negative recall 6.32 (5.20) 0.85 (0.90) 5.00 (4.00) 10.40 (3.44) <.001 AD vs. HE, p < .001, d = 3.80
Neutral recall 5.00 (4.66) 0.77 (0.93) 5.25 (5.23) 7.65 (3.94) <.001 AD vs. HE, p < .002, d = 2.40
Positive recognition 14.51 (3.48) 11.38 (2.66) 15.00 (2.33) 16.35 (2.94) <.001 AD vs. HE, p < .001, d = 1.77
Negative recognition 16.07 (3.66) 12.15 (2.82) 16.50 (2.73) 18.45 (1.99) <.001 AD vs. HE, p < .001, d = 2.58
Neutral recognition 15.39 (3.73) 11.85 (2.94) 15.75 (3.58) 17.55 (2.35) <.001 AD vs. HE, p < .001, d = 2.14
IGT net score 16.15 (29.88) 0.77 (18.86) 20.00 (33.92) 24.60 (31.42) .104 _
IGT shifting decks 55.61 (26.45) 70.31 (16.03) 58.25 (33.69) 45.00 (24.97) <.05 AD vs. HE, p < .05, d = −1.21
BART 25.39 (16.20) 23.93 (16.80) 26.79 (20.32) 25.78 (14.81) .943 _
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment; HE = healthy elderly; UPSIT = University of Pennsylvania
Smell Identification Test; OMT = Odor Memory Test; IGT = Iowa Gambling Task; BART = Balloon Analogue Risk Task. The interpretation of the effect size is
as follows: 0.2 = small, 0.5 = medium, and 0.8 = large.

Table 3. Pearson correlations between olfaction and higher cortical functions in healthy older adults, MCI, and AD.
Recall Recognition
Variables Positive Negative Neutral Positive Negative Neutral IGT net score IGT shifting decks BART WTAR
Odor identification .669*** .768*** .698*** .424*** .612*** .534*** .177 −.344* −.051 .334*
Odor memory .611*** .643*** .566*** .373* .410** .306 .148 −.357* .154 .213
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment; IGT = Iowa Gambling Task; BART = Balloon Analogue Risk
Task; WTAR = Wechsler Test of Adult Reading.
*p < .05. **p < .01. ***p < .001.
652 A. M. WARD ET AL.

Table 4. Hierarchical multiple regression model: Odor memory a statistically significant increase in R2 of .101, F(1,
and positive recall in healthy older adults, MCI, and AD. 38) = 11.684, p < .01 (Table 5). The addition of odor
Positive recall memory to the prediction of negative recall led to a
Model 1 Model 2
statistically significant increase of R2 of .071, F(1,
Variable B β p B β p
38) = 7.483, p < .01 (Table 6).
Constant −4.289 .014* −4.582 .007**
General cognition 0.538 .747 <.001*** 0.425 .590 <.001*** Hierarchical regression models were completed for
Odor memory 0.464 .281 .027* each olfactory variable as an independent variable and
F 49.376 <.001*** 30.065 <.001***
R2 .559 .613 positive recognition and negative recognition as each
ΔF 49.376 <.001*** 5.304 .027* outcome variable. For both the odor identification and
ΔR2 .559 .054
odor memory models, general cognition was significant
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild
cognitive impairment. when it was entered into each first block [positive
*p < .05. **p < .01. ***p < .001. recognition: F(1, 39) = 6.070, p < .05; negative recogni-
tion: F(1, 39) = 47.344, p < .001]. The addition of
olfaction to the prediction of both positive recognition
hierarchical regression model. General cognition was and negative recognition did not lead to a statistically
significant when it was entered into the first block in significant increase in R2 for the odor identification
each model, F(1, 39) = 49.376, p < .001. The addition of models or the odor memory models (p > .05).
odor identification to the prediction of positive recall
did not lead to a statistically significant increase in R2
Relationship between olfaction and neutral
(p > .05). The addition of odor memory to the predic-
memory
tion of positive recall led to a statistically significant
increase in R2 of .054, F(1, 38) = 5.304, p < .05 (Table 4). Using Pearson correlations, odor identification and
Hierarchical regression models were completed for odor memory were significantly associated with neutral
each olfactory variable as the independent variable and recall (p < .001; Table 3). Odor identification was sig-
negative recall as the outcome variable. General cogni- nificantly associated with neutral recognition
tion was significant when it was entered into the first (p < .001), but odor memory was not (p > .05; Table 3).
block, F(1, 39) = 51.666, p < .001. The addition of odor Hierarchical regression models were completed for
identification to the prediction of negative recall led to each olfactory variable as the independent variables and

Table 5. Hierarchical multiple regression model: Odor identification and negative recall in healthy older adults, MCI, and AD.
Negative recall
Model 1 Model 2
Variable B β p B β p
Constant −4.970 .005** −6.260 <.001***
General cognition 0.546 .755 <.001*** 0.301 .416 .004**
Odor identification 0.245 .465 .002**
F 51.666 <.001*** 38.752 <.001***
R2 .570 .671
ΔF 51.666 <.001*** 11.684 .002**
ΔR2 .570 .101
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment.
**p < .01. ***p < .001.

Table 6. Hierarchical multiple regression model: Odor memory and negative recall in healthy older adults, MCI, and AD.
Negative recall
Model 1 Model 2
Variable B β p B β p
Constant −4.970 .005** −5.306 .001**
General cognition 0.546 .755 <.001*** 0.416 .575 <.001***
Odor memory 0.534 .321 .009**
F 51.666 <.001*** 33.868 <.001***
R2 .570 .641
ΔF 51.666 <.001*** 7.483 .009**
ΔR2 .570 .071
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment.
**p < .01. ***p < .001.
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 653

neutral recall as the outcome variable. A hierarchical associated with IGT net score or BART (p > .05;
regression model was also completed for odor identifica- Table 3).
tion as the independent variable and neutral recognition as Hierarchical regression models were completed for
the outcome variable. General cognition was entered into each olfactory variable as independent variables and shift-
the first block of each hierarchical regression model. In the ing between decks as the outcome variable. General cog-
neutral recall models, general cognition was significant nition was entered into the first block of each hierarchical
when it was entered into the first block, F(1, regression model. General cognition was significant when
39) = 29.491, p < .001. The addition of odor identification it was entered into the first block, F(1, 39) = 10.857,
to the prediction of neutral recall led to a statistically p < .01. The addition of each olfactory variable to the
significant increase in R2 of .103, F(1, 38) = 8.374, p < .01 prediction of shifting between decks did not lead to a
(Table 7). The addition of odor memory to the prediction statistically significant increase in R2 (p > .05).
of neutral recall led to a statistically significant increase in
R2 of .057, F(1, 38) = 4.255, p < .05 (Table 8). In the neutral
recognition model for odor identification, general cogni-
Relationship between olfaction and reading
tion was significant when it was entered into the first block,
F(1, 39) = 14.967, p < .001. The addition of odor identifica- Using Pearson correlations, odor identification was
tion to the prediction of neutral recognition did not lead to significantly associated with WTAR (p < .05), while
a statistically significant increase in R2 (p > .05). odor memory was not (p > .05; Table 3). The WTAR
was administered as a variable that we did not expect to
relate to olfaction over and above general cognition.
General cognition was entered into the first block of
Relationship between olfaction and decision making
the hierarchical regression model. General cognition
Using Pearson correlations, odor identification and was significant when it was entered into the first
odor memory were significantly associated with shift- blocks, F(1, 39) = 21.442, p < .001. The addition of
ing between decks in the IGT (p < .05; Table 3). Odor odor identification to the prediction of reading did not
identification and odor memory were not significantly lead to a statistically significant increase in R2 (p > .05).

Table 7. Hierarchical multiple regression model: Odor identification and neutral recall in healthy older adults, MCI, and AD.
Neutral recall
Model 1 Model 2
Variable B β p B β p
Constant −3.794 .033* −4.960 .004**
General cognition 0.426 .656 <.001*** 0.204 .315 .059
Odor identification 0.222 .468 .006**
F 29.491 <.001*** 21.721 <.001***
R2 .431 .533
ΔF 29.491** <.001*** 8.374 .006***
ΔR2 .431 .103
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment.
*p < .05. **p < .01. ***p < .001.

Table 8. Hierarchical multiple regression model: Odor memory and neutral recall in healthy older adults, MCI, and AD.
Neutral recall
Model 1 Model 2
Variable B β p B β p
Constant −3.794 .033* −4.065 .018*
General cognition 0.426 .656 <.001*** 0.321 .495 .001**
Odor memory 0.431 .289 .046*
F 29.491 <.001*** 18.104 <.001***
R2 .431 .488
ΔF 29.491** <.001*** 4.255 .046*
ΔR2 .431 .057
Note. AD = patients with Alzheimer’s disease; MCI = participants with mild cognitive impairment.
*p < .05. **p < .01. ***p < .001.
654 A. M. WARD ET AL.

Comparing the strength of the association between in the primary olfactory cortex, and then projections
olfaction and emotional memory and olfaction and are sent to proximal neural regions important for
neutral memory memory, including the anterior medial temporal lobe
(AMTL; Hawkes, 2009). Olfaction and memory are
An analysis was conducted to determine whether the
likely related because both involve neural processing
relationship between olfaction and emotional memory
in the hippocampus, which is responsible for binding
was stronger than the relationship between olfaction
together elements of an experience, including sensory
and neutral memory.
information, to form memories (Eichenbaum, 1998;
In all eight comparisons, there were no significant
Goodrich-Hunsaker, Gilbert, & Hopkins, 2009;
differences between the strength of the association
Konkel & Cohen, 2009). Olfactory functioning may
between olfaction and emotional memory and the
measure the neural integrity of the AMTL, including
strength of the association between olfaction and neu-
the hippocampus and amygdala, especially since pre-
tral memory (p < .05).
vious research suggests that neuropathology is present
in these regions in patients with AD (Arnold, Hyman,
Flory, Damasio, & Van Hoesen, 1991) and MCI
Discussion
(Markesbery, 2010). However, this cannot be con-
The relationship between olfaction and higher cortical firmed in this study because data regarding neural
functions was investigated in healthy older adults, integrity in these patients were not available.
MCI, and AD. Overall, a significant association was The current study did not find a significant difference
found between olfaction (using odor identification between the relationship between olfaction and emo-
and odor memory) and both recall and recognition. tional memory and the relationship between olfaction
However, a stronger association was found for recall and nonemotional memory. We completed a follow-up
since the relationship between each olfactory variable analysis to verify that the lack of findings were not due to
and memory recall was significant over and above emotional stimuli that did not appropriately induce
general cognition. Previous research found a close rela- more arousal than neutral stimuli. There was a signifi-
tionship between olfaction and neuropsychological cant difference with a greater arousal ratings for emo-
measures of memory recall (Devanand et al., 2010; tional memory stimuli (p < .05). However, it is still
Djordjevic et al., 2008; Fusetti et al., 2010; Lehrner possible that olfaction is more strongly related to per-
et al., 2009; Makizako et al., 2014; Royall et al., 2002) sonally relevant memories since they tend to induce
and found that olfaction is an effective cue for auto- stronger emotions than those induced by non-person-
biographical memory (Chu & Downes, 2002; Herz, ally relevant stimuli. This would support previous
2004; Herz et al., 2004; Willander & Larsson, 2007). research that has shown that olfactory cues induce emo-
This study supports the relationship between each tional autobiographical memories (Herz, 1998; Herz
olfactory variable and neuropsychological measures of et al., 2004; Willander & Larsson, 2007) and are better
memory. It also extends these findings to emotional than other sensory stimuli at doing so (Herz, 1998).
memory and confirms that the relationship between We found a stronger relationship between each
olfaction and memory is stronger with memory recall olfactory variable and memory recall than between
rather than with recognition. The association between each olfactory variable and memory recognition.
odor identification and reading ability was significant. These findings are consistent with many studies that
Although the WTAR is used to measure premorbid suggest a significant relationship between olfaction and
functioning, performance has been shown to decrease memory recall using neuropsychological tasks
in those with dementia, so this finding may reflect this (Devanand et al., 2010; Djordjevic et al., 2008;
decline (McFarlane et al., 2006). The association Hedner, Larsson, Arnold, Zucco, & Hummel, 2010;
between odor identification and reading ability was Lehrner et al., 2009; Makizako et al., 2014; Royall
not significant over and above general cognition, sug- et al., 2002). Few studies have examined the relation-
gesting that the relationship between olfaction and ship between olfaction and memory recognition and
memory recall is not only due to overall cognitive found no significant association between olfaction and
decline in aging and neurodegenerative conditions. memory recognition (Murphy, Nordin, & Acosta,
The significant relationship between each olfactory 1997) or found a significant association that was
variable and both nonemotional and emotional mem- small in effect size (odor threshold r = .173; odor
ory supports previous research that suggests a close discrimination r = .202; odor identification r = .160;
connection in neural processing. Previous research Hedner et al., 2010). While there is no clear explana-
has shown that olfactory information is first processed tion for the weaker association between olfaction and
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 655

recognition memory, we would like to emphasize that than the BART since they involve estimating the prob-
free recall and recognition are not types of memory, ability of an outcome, which is more closely related to
per se, but, rather, methods to test memory. Different value-based decision making (Clark et al., 2008; Clark
methods can lead to different results, needless to say, & Manes, 2004).
depending on the robustness of the memories and There were some limitations to our study. Although
other related factors. Different outcomes for free recall our inclusion of participants who varied widely in cogni-
versus recognition formats are not uncommon in the tive and olfactory abilities is a strength in determining the
literature, especially for persons with memory pro- relative importance of olfaction to higher cortical func-
blems, and this could help explain our findings. tioning, our sample was modest. Future research should
We found a significant relationship between each collect more data in each patient subgroup (e.g., MCI,
olfactory variable and one measure of decision making, AD) when exploring olfaction and higher cognitive func-
but each olfactory variable was not significantly related tioning. This work may elucidate whether the relation-
to decision making over and above general cognition. ships differ amongst different groups. Our goal in the
This is contradictory to previous research that has current study, however, was to make an initial foray into
found a relationship between olfaction and executive this topic by exploring olfaction–cognitive associations
functioning in healthy older adults, MCI, and AD generally, and for this initial effort, a broader approach
(Djordjevic et al., 2008; Lehrner et al., 2009; Makizako with more range in the variables was considered optimal.
et al., 2014; Royall et al., 2002). However, these studies Some of patients with AD and MCI were recruited from
mainly used tasks measuring working memory, shift- another institution without access to a neurocognitive
ing, and attention, which capture different cognitive exam or in-depth history. Furthermore, neuroimaging
abilities than our decision-making tasks. Furthermore, data and comprehensive neuropsychological data were
we had expected that including patients with AD and not available on participants to fully characterize their
MCI and healthy older adults would allow us to inves- cognitive and neural dysfunction.
tigate our questions of interest since these populations There are several areas of future research that could
have a wide spectrum of olfactory performances, cog- be addressed based on these findings. Future research
nitive abilities, and underlying neuropathology. could investigate whether the relationship between
Patients with MCI and AD share similar neuropathol- olfaction and emotional memory is present in other
ogy, such as plaques and tangles, and neuropathology is neurological populations, or if it is specific to neural
more marked in patients with AD (Bennett, Schneider, decline in the medial temporal lobe. This relationship
Bienias, Evans, & Wilson, 2005). Studies in patients could be explored in Parkinson’s patients, since they
with MCI show evidence of atrophy in regions outside also have documented olfactory impairments (Doty,
the medial temporal lobe, such as frontal regions 2012) and have mixed findings regarding the relation-
(Hämäläinen et al., 2007; Singh et al., 2006; Whitwell ship between olfaction and cognition (Bohnen et al.,
et al., 2008), and the orbitofrontal cortex is then more 2010; Doty, Riklan, Deems, Reynolds, & Stellar, 1989;
strongly affected when MCI transitions to AD (Braak & Postuma & Gagnon, 2010). Investigation of the olfac-
Bra Ak, 1991). Since there were not significant findings tory deficits in other neurological populations may
surrounding decision-making abilities in this study, it reveal whether the relationship between olfaction and
is possible that neural decline was more localized to the emotional memory is unique to MCI and AD
medial temporal lobe than to the orbitofrontal cortex. populations.
This was supported by the current findings that there Based on the results of this study, olfaction may be
was no significant difference between healthy older an indicator of memory recall over and above general
adults, MCI, and AD in two of the decision-making cognition. These findings are especially relevant to
variables; however, this cannot be confirmed in this patients with MCI and AD since impairment occurs
study since no neuroimaging data were available. early in each of these domains. Exploration of olfactory
In this study, we found that patients with AD had processing deficits in neurodegenerative disease may
impaired value-based decision making as indicated by provide a tool to assess cognitive functioning.
higher shifting between decks, which replicated pre-
vious findings (Sinz et al., 2008). However, risk-taking
behavior was not impaired in patients with MCI or AD Acknowledgments
using the BART. Previous studies have found impaired
The authors thank members of the Benton Neuropsychology
risk-taking behavior in patients with AD (Delazer et al., Laboratory and the Alzheimer’s Association for help with
2007; Sinz et al., 2008) and MCI (Zamarian et al., recruitment. We also thank Kiwanis International,
2011). However, these risk-taking tasks are different McDonnell Foundation, and Fraternal Order of Eagles.
656 A. M. WARD ET AL.

Disclosure statement in humans. Psychoneuroendocrinology, 26(3), 307–317.

doi:10.1016/S0306-4530(00)00058-5
No potential conflict of interest was reported by the authors. Buchanan, T. W., Tranel, D., & Adolphs, R. (2003). A spe-
cific role for the human amygdala in olfactory memory.
Learning & Memory, 10(5), 319–325. doi:10.1101/
Funding lm.62303
This research was supported by Kiwanis International; the Choudhury, E. S., Moberg, P., & Doty, R. L. (2003).
James S. McDonnell Foundation [grant number McDonnell Influences of age and sex on a microencapsulated odor
UHC-Collab 220020387]; and the Fraternal Order of Eagles. memory test. Chemical Senses, 28(9), 799–805.
Chu, S., & Downes, J. J. (2002). Proust nose best: Odors are
better cues of autobiographical memory. Memory &
Cognition, 30(4), 511–518. doi:10.3758/BF03194952
References Clark, L., Bechara, A., Damasio, H., Aitken, M., Sahakian, B.,
& Robbins, T. (2008). Differential effects of insular and
Albert, M. S., DeKosky, S. T., Dickson, D., Dubois, B.,
ventromedial prefrontal cortex lesions on risky decision-
Feldman, H. H., Fox, N. C., . . . Phelps, C. H. (2011). The
making. Brain, 131, 1311–1322. doi:10.1093/brain/awn066
diagnosis of mild cognitive impairment due to Alzheimer’s
Clark, L., & Manes, F. (2004). Social and emotional decision-
disease: Recommendations from the National Institute on
making following frontal lobe injury. Neurocase, 10(5),
Aging-Alzheimer’s Association workgroups on diagnostic
398–403. doi:10.1080/13554790490882799
guidelines for Alzheimer’s disease. Alzheimer’s &
Cohen, J. (1988). Statistical power analysis for the behavioral
Dementia, 7(3), 270–279. doi:10.1016/[Link].2011.03.008
sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R., & Van
Conti, M. Z., Vicini-Chilovi, B., Riva, M., Zanetti, M.,
Hoesen, G. W. (1991). The topographical and neuroana-
Liberini, P., Padovani, A., & Rozzini, L. (2013). Odor
tomical distribution of neurofibrillary tangles and neuritic
identification deficit predicts clinical conversion from
plaques in the cerebral cortex of patients with Alzheimer’s
mild cognitive impairment to dementia due to
disease. Cerebral Cortex, 1(1), 103–116. doi:10.1093/cer-
Alzheimer’s disease. Archives of Clinical Neuropsychology,
cor/1.1.103
28(5), 391–399. doi:10.1093/arclin/act032
Bahar-Fuchs, A., Moss, S., Rowe, C., & Savage, G. (2011).
Dade, L. A., Jones-Gotman, M., Zatorre, R. J., & Evans, A. C.
Awareness of olfactory deficits in healthy aging, amnestic
(1998). Human brain function during odor encoding and
mild cognitive impairment and Alzheimer’s disease.
recognition: A PET activation study. Annals of the New
International Psychogeriatrics, 23(07), 1097–1106.
York Academy of Sciences, 855, 572–574. doi:10.1111/
doi:10.1017/S1041610210002371
[Link]-1
Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion,
Delazer, M., Sinz, H., Zamarian, L., & Benke, T. (2007).
decision making, and the orbitofrontal cortex. Cerebral
Decision-making with explicit and stable rules in mild
Cortex, 10(3), 295–307. doi:10.1093/cercor/10.3.295
Alzheimer’s disease. Neuropsychologia, 45(8), 1632–1641.
Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R.
doi:10.1016/[Link].2007.01.006
(1997). Deciding advantageously before knowing the
DeRenzo, E. G., Conley, R. R., & Love, R. (1998). Assessment
advantageous strategy. Science, 275(5304), 1293–1295.
of capacity to give consent to research participation: State-
doi:10.1126/science.275.5304.1293
of-the-art and beyond. Journal of Health Care Law and
Beck, A., Steer, R., & Brown, G. (1996). Beck depression
Policy, 1(1), 66–87.
inventory-II manual. San Antonio, TX: Psychological
Devanand, D. P., Lee, S., Manly, J., Andrews, H., Schupf, N.,
Corporation.
Doty, R. L., . . . Mayeux, R. (2015). Olfactory deficits pre-
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., &
dict cognitive decline and Alzheimer dementia in an urban
Erbaugh, J. (1961). An inventory for measuring depres-
community. Neurology, 84(2), 182–189. doi:10.1212/
sion. Archives of General Psychiatry, 4, 561–571.
WNL.0000000000001132
doi:10.1001/archpsyc.1961.01710120031004
Devanand, D. P., Michaels-Marston, K. S., Liu, X., Pelton, G.
Bennett, D., Schneider, J., Bienias, J., Evans, D., & Wilson, R.
H., Padilla, M., Marder, K., . . . Mayeux, R. (2000).
(2005). Mild cognitive impairment is related to Alzheimer
Olfactory deficits in patients with mild cognitive impair-
disease pathology and cerebral infarctions. Neurology, 64
ment predict Alzheimer’s disease at follow-up. The
(5), 834–841. doi:10.1212/[Link].0000152982.47274.9E
American Journal of Psychiatry, 157(9), 1399–1405.
Bohnen, N. I., Muller, M. L., Kotagal, V., Koeppe, R. A.,
doi:10.1176/[Link].157.9.1399
Kilbourn, M. A., Albin, R. L., & Frey, K. A. (2010).
Devanand, D. P., Tabert, M. H., Cuasay, K., Manly, J. J.,
Olfactory dysfunction, central cholinergic integrity and
Schupf, N., Brickman, A. M., . . . Mayeux, R. (2010).
cognitive impairment in Parkinson’s disease. Brain, 133,
Olfactory identification deficits and MCI in a multi-ethnic
1747–1754. doi:10.1093/brain/awq079
elderly community sample. Neurobiology of Aging, 31(9),
Braak, H., & Bra Ak, E. (1991). Neuropathological stageing of
1593–1600. doi:10.1016/[Link].2008.09.008
Alzheimer-related changes. Acta Neuropathologica, 82(4),
Djordjevic, J., Jones-Gotman, M., De Sousa, K., & Chertkow, H.
239–259. doi:10.1007/BF00308809
(2008). Olfaction in patients with mild cognitive impairment
Buchanan, T. W. (2007). Retrieval of emotional memories.
and Alzheimer’s disease. Neurobiology of Aging, 29(5), 693–
Psychological Bulletin, 133(5), 761–779. doi:10.1037/0033-
706. doi:10.1016/[Link].2006.11.014
2909.133.5.761
Dolcos, F., LaBar, K. S., & Cabeza, R. (2004). Interaction
Buchanan, T. W., & Lovallo, W. R. (2001). Enhanced memory
between the amygdala and the medial temporal lobe
for emotional material following stress-level cortisol treatment
 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 657

memory system predicts better memory for emotional of Clinical and Experimental Neuropsychology, 32(10),
events. Neuron, 42(5), 855–863. doi:10.1016/S0896-6273 1062–1067. doi:10.1080/13803391003683070
(04)00289-2 Herz, R. S. (1998). Are odors the best cues to memory? A cross-
Doty, R. L. (1983). The smell identification test administration modal comparison of associative memory stimuli. Annals of
manual. Haddon Heights, NJ: Sensonics. the New York Academy of Sciences, 855, 670–674.
Doty, R. L. (2003). The odor memory test administration doi:10.1111/[Link]-1
manual (2nd ed.). Haddon Heights, NJ: Sensonics. Herz, R. S. (2004). A naturalistic analysis of autobiographical
Doty, R. L. (2012). Olfactory dysfunction in Parkinson dis- memories triggered by olfactory visual and auditory sti-
ease. Nature Reviews Neurology, 8(6), 329–339. muli. Chemical Senses, 29(3), 217–224. doi:10.1093/
doi:10.1038/nrneurol.2012.80 chemse/bjh025
Doty, R. L., McKeown, D. A., Lee, W. W., & Shaman, P. Herz, R. S., Eliassen, J., Beland, S., & Souza, T. (2004).
(1995). A study of the test-retest reliability of ten olfactory Neuroimaging evidence for the emotional potency of
tests. Chemical Senses, 20(6), 645–656. doi:10.1093/ odor-evoked memory. Neuropsychologia, 42(3), 371–378.
chemse/20.6.645 doi:10.1016/[Link].2003.08.009
Doty, R. L., Reyes, P. F., & Gregor, T. (1987). Presence of Jiang, R. S., Lu, F. J., Liang, K. L., Shiao, J. Y., Su, M. C., Hsin,
both odor identification and detection deficits in C. H., & Chen, W. K. (2008). Olfactory function in
Alzheimer’s disease. Brain Research Bulletin, 18(5), 597– patients with chronic rhinosinusitis before and after func-
600. doi:10.1016/0361-9230(87)90129-8 tional endoscopic sinus surgery. American Journal of
Doty, R. L., Riklan, M., Deems, D. A., Reynolds, C., & Stellar, Rhinology, 22(4), 445–448. doi:10.2500/ajr.2008.22.3195
S. (1989). The olfactory and cognitive deficits of Kensinger, E. A., Brierley, B., Medford, N., Growdon, J. H., &
Parkinson’s disease: Evidence for independence. Annals of Corkin, S. (2002). Effects of normal aging and Alzheimer’s
Neurology, 25(2), 166–171. doi:10.1002/(ISSN)1531-8249 disease on emotional memory. Emotion, 2(2), 118–134.
Doty, R. L., Kisat, M., & Tourbier, I. (2008). Estrogen repla- doi:10.1037/1528-3542.2.2.118
cement therapy induces functional asymmetry on an odor Kesslak, J. P., Cotman, C. W., Chui, H. C., Van Den Noort,
memory/discrimination test. Brain Research, 1214, 35–39. S., Fang, H., Pfeffer, R., & Lynch, G. (1988). Olfactory tests
doi:10.1016/[Link].2008.04.017 as possible probes for detecting and monitoring
Eibenstein, A., Fioretti, A., Simaskou, M., Sucapane, P., Alzheimer’s disease. Neurobiology of Aging, 9(4), 399–
Mearelli, S., Mina, C., . . . Fusetti, M. (2005). Olfactory 403. doi:10.1016/S0197-4580(88)80087-3
screening test in mild cognitive impairment. Neurological Konkel, A., & Cohen, N. J. (2009). Relational memory and the
Sciences, 26(3), 156–160. doi:10.1007/s10072-005-0453-2 hippocampus: Representations and methods. Frontiers in
Eichenbaum, H. (1998). Using olfaction to study memory. Neuroscience, 3, 166–174. doi:10.3389/neuro.01.023.2009
Annals of the New York Academy of Sciences, 855(1), 657– Koss, E. (1986). Olfactory dysfunction in Alzheimer’s disease.
669. doi:10.1111/[Link]-1 Developmental Neuropsychology, 2(2), 89–99. doi:10.1080/
Fujie, S., Namiki, C., Nishi, H., Yamada, M., Miyata, J., 87565648609540332
Sakata, D., . . . Murai, T. (2008). The role of the uncinate Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008).
fasciculus in memory and emotional recognition in International affective picture system (IAPS): Affective rat-
amnestic mild cognitive impairment. Dementia and ings of pictures and instruction manual (Technical Report
Geriatric Cognitive Disorders, 26(5), 432–439. A-8). Gainesville, FL: University of Florida.
doi:10.1159/000165381 Lehrner, J., Pusswald, G., Gleiss, A., Auff, E., & Dal-Bianco,
Fusetti, M., Fioretti, A. B., Silvagni, F., Simaskou, M., P. (2009). Odor identification and self-reported olfactory
Sucapane, P., Necozione, S., & Eibenstein, A. (2010). functioning in patients with subtypes of mild cognitive
Smell and preclinical Alzheimer disease: Study of 29 impairment. The Clinical Neuropsychologist, 23(5), 818–
patients with amnesic mild cognitive impairment. Journal 830. doi:10.1080/13854040802585030
of Otolaryngology Head & Neck Surgery, 39(2), 175–181. Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B.,
Goodrich-Hunsaker, N. J., Gilbert, P. E., & Hopkins, R. O. Ramsey, S. E., Stuart, G. L., . . . Brown, R. A. (2002).
(2009). The role of the human hippocampus in odor–place Evaluation of a behavioral measure of risk taking: The
associative memory. Chemical Senses, 34(6), 513–521. Balloon Analogue Risk Task (BART). Journal of
doi:10.1093/chemse/bjp026 Experimental Psychology Applied, 8(2), 75–84.
Hämäläinen, A., Tervo, S., Grau-Olivares, M., Niskanen, E., doi:10.1037/1076-898X.8.2.75
Pennanen, C., Huuskonen, J., . . . Soininen, H. (2007). Lojkowska, W., Sawicka, B., Gugala, M., Sienkiewicz-Jarosz,
Voxel-based morphometry to detect brain atrophy in pro- H., Bochynska, A., Scinska, A., . . . Ryglewicz, D. (2011).
gressive mild cognitive impairment. Neuroimage, 37(4), Follow-up study of olfactory deficits, cognitive functions,
1122–1131. doi:10.1016/[Link].2007.06.016 and volume loss of medial temporal lobe structures in
Hamann, S. B., Monarch, E. S., & Goldstein, F. C. (2000). patients with mild cognitive impairment. Current
Memory enhancement for emotional stimuli is impaired in Alzheimer Research, 8(6), 689–698. doi:10.2174/
early Alzheimer’s disease. Neuropsychology, 14(1), 82–92. 156720511796717212
doi:10.1037/0894-4105.14.1.82 Makizako, M., Makizako, H., Doi, T., Uemura, K.,
Hawkes, C. H. (2009). The neurology of olfaction. Cambridge: Tsutsumimoto, K., Miyaguchi, H., & Shimada, H. (2014).
Cambridge University Press. Olfactory identification and cognitive performance in
Hedner, M., Larsson, M., Arnold, N., Zucco, G. M., & community-dwelling older adults with mild cognitive
Hummel, T. (2010). Cognitive factors in odor detection, impairment. Chemical Senses, 39(1), 39–46. doi:10.1093/
odor discrimination, and odor identification tasks. Journal chemse/bjt052
658 A. M. WARD ET AL.

Makowska, I., Kloszewska, I., Grabowska, A., Szatkowska, I., & Savic, I. (2002). Imaging of brain activation by odorants in
Rymarczyk, K. (2011). Olfactory deficits in normal aging humans. Current Opinion in Neurobiology, 12(4), 455–461.
and Alzheimer’s disease in the polish elderly population. doi:10.1016/S0959-4388(02)00346-X
Archives of Clinical Neuropsychology, 26(3), 270–279. Singh, V., Chertkow, H., Lerch, J. P., Evans, A. C., Dorr, A.
doi:10.1093/arclin/acr011 E., & Kabani, N. J. (2006). Spatial patterns of cortical
Markesbery, W. R. (2010). Neuropathologic alterations in thinning in mild cognitive impairment and Alzheimer’s
mild cognitive impairment: A review. Journal of disease. Brain, 129(11), 2885–2893. doi:10.1093/brain/
Alzheimer’s Disease, 19(1), 221–228. awl256
McFarlane, J., Welch, J., & Rodgers, J. (2006). Severity of Sinz, H., Zamarian, L., Benke, T., Wenning, G. K., & Delazer,
Alzheimer’s disease and effect on premorbid measures of M. (2008). Impact of ambiguity and risk on decision
intelligence. British Journal of Clinical Psychology, 45(4), making in mild Alzheimer’s disease. Neuropsychologia, 46
453–464. doi:10.1348/014466505X71245 (7), 2043–2055. doi:10.1016/j.
McKhann, G. M., Knopman, D. S., Chertkow, H., Hyman, B. neuropsychologia.2008.02.002
T., Jack, C. R., Kawas, C. H., . . . Phelps, C. H. (2011). The Strenziok, M., Pulaski, S., Krueger, F., Zamboni, G., Clawson,
diagnosis of dementia due to Alzheimer’s disease: D., & Grafman, J. (2011). Regional brain atrophy and
Recommendations from the National Institute on Aging- impaired decision making on the balloon analog risk
Alzheimer’s Association workgroups on diagnostic guide- task in behavioral variant frontotemporal dementia.
lines for Alzheimer’s disease. Alzheimer’s & Dementia, 7 Cognitive and Behavioral Neurology, 24(2), 59–67.
(3), 263–269. doi:10.1016/[Link].2011.03.005 doi:10.1097/WNN.0b013e3182255a7c
Murphy, C., Gilmore, M. M., Seery, C. S., Salmon, D. P., & Tourbier, I. A., & Doty, R. L. (2007). Sniff magnitude test:
Lasker, B. R. (1990). Olfactory thresholds are associated Relationship to odor identification, detection, and mem-
with degree of dementia of Alzheimer’s disease. ory tests in a clinic population. Chemical Senses, 32(6),
Neurobiology of Aging, 11(4), 465–469. doi:10.1016/0197- 515–523. doi:10.1093/chemse/bjm020
4580(90)90014-Q Vyhnalek, M., Magerova, H., Andel, R., Nikolai, T.,
Murphy, C., Nordin, S., & Acosta, L. (1997). Odor learning, Kadlecova, A., Laczo, J., & Hort, J. (2015). Olfactory iden-
recall, and recognition memory in young and elderly tification in amnestic and non-amnestic mild cognitive
adults. Neuropsychology, 11(1), 126–137. doi:10.1037/ impairment and its neuropsychological correlates.
0894-4105.11.1.126 Journal of the Neurological Sciences, 349(1–2), 179–184.
Murphy, C., Schubert, C. R., Cruickshanks, K. J., Klein, B. E., doi:10.1016/[Link].2015.01.014
Klein, R., & Nondahl, D. M. (2002). Prevalence of olfac- Wang, J., Eslinger, P. J., Smith, M. B., & Yang, Q. X. (2005).
tory impairment in older adults. The Journal of the Functional magnetic resonance imaging study of human
American Medical Association, 288(18), 2307–2312. olfaction and normal aging. The Journals of Gerontology
doi:10.1001/jama.288.18.2307 Series A: Biological Sciences and Medical Sciences, 60(4),
Nasreddine, Z. S., Phillips, N. A., Bedirian, V., Charbonneau, S., 510–514. doi:10.1093/gerona/60.4.510
Whitehead, V., Collin, I., . . . Chertkow, H. (2005). The Warner, M. D., Peabody, C. A., Flattery, J. J., & Tinklenberg,
Montreal Cognitive Assessment, MoCA: A brief screening J. R. (1986). Olfactory deficits and Alzheimer’s disease.
tool for mild cognitive impairment. Journal of the American Biological Psychiatry, 21(1), 116–118. doi:10.1016/0006-
Geriatrics Society, 53(4), 695–699. doi:10.1111/jgs.2005.53. 3223(86)90013-2
issue-4 Wechsler, D. (2001). Wechsler Test of Adult Reading: WTAR.
Naudin, M., Mondon, K., El-Hage, W., Desmidt, T., Jaafari, N., San Antonio, TX: Psychological Corporation.
Belzung, C., . . . Atanasova, B. (2014). Long-term odor recog- Whitwell, J. L., Shiung, M. M., Przybelski, S., Weigand, S. D.,
nition memory in unipolar major depression and Alzheimer‫׳‬s Knopman, D. S., Boeve, B. F., . . . Jack, C. R. (2008). MRI
disease. Psychiatry Research, 220(3), 861–866. doi:10.1016/j. patterns of atrophy associated with progression to AD in
psychres.2014.08.050 amnestic mild cognitive impairment. Neurology, 70(7),
Nordin, S., Monsch, A. U., & Murphy, C. (1995). Unawareness 512–520. doi:10.1212/[Link].0000280575.77437.a2
of smell loss in normal aging and Alzheimer’s disease: Willander, J., & Larsson, M. (2007). Olfaction and emotion:
Discrepancy between self-reported and diagnosed smell The case of autobiographical memory. Memory &
sensitivity. The Journals of Gerontology Series B: Cognition, 35(7), 1659–1663. doi:10.3758/BF03193499
Psychological Sciences and Social Sciences, 50B(4), P187– Wilson, R. S., Schneider, J. A., Arnold, S. E., Tang, Y., Boyle,
P192. doi:10.1093/geronb/50B.4.P187 P. A., & Bennett, D. A. (2007). Olfactory identification and
Postuma, R., & Gagnon, J. (2010). Cognition and olfaction in incidence of mild cognitive impairment in older age.
Parkinson’s disease [Letter to the editor]. Brain, 133, Archives of General Psychiatry, 64(7), 802–808.
e160–e160. doi:10.1093/brain/awq225 doi:10.1001/archpsyc.64.7.802
Rey, A. (1958). L’examen clinique en psychologie. Zamarian, L., Weiss, E. M., & Delazer, M. (2011). The impact
Royall, D. R., Chiodo, L. K., Polk, M. J., & Jaramillo, C. J. (2002). of mild cognitive impairment on decision making in two
Severe dysosmia is specifically associated with Alzheimer- gambling tasks. The Journals of Gerontology Series B:
like memory deficits in nondemented elderly retirees. Psychological Sciences and Social Sciences, 66B(1), 23–31.
Neuroepidemiology, 21(2), 68–73. doi:10.1159/000048619 doi:10.1093/geronb/gbq067
Copyright of Journal of Clinical & Experimental Neuropsychology is the property of Taylor
& Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.