Chem. Senses 24: 337-346,
1999
© Oxford University Press
Odor Identification, Consistency of Label Use, Olfactory Threshold and their Relationships to Odor Memory over the Human Lifespan
Neurologische Universitätsklinik Institut für Psychologie, Universität Wien, Wien, Austria; 1 Abteilung für Entwicklungspsychologie Institut für Psychologie, Universität Wien, Wien, Austria 2 Institut für Medizinische Psychologie, Ludwig-Maximilian-Universität München, München, Germany
Mag. Johann Lehrner, Neurologische Universitätsklinik, Allgemeines Krankenhaus, Währingergürtel 18–20, A-1097 Wien, Austria. e-mail:hannes.lerner{at}akh-wien.ac.at
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The purpose of this study was to investigate olfactory threshold, odor identification, consistency of label use and their relationships to odor memory in the context of semantic/episodic memory across the human lifespan. A total of 137 subjects aged 4–90 years were tested with several olfactory test procedures. We found that olfactory sensitivity was well developed in children despite the finding that their odor naming and odor memory were inferior to that of adults. In the elderly population, olfactory functions gradually declined, with odor memory and odor identification demonstrating the most significant decline. Semantic encoding was differentially related to odor memory over the human age span. Whereas consistency of label use was the main predictor for odor memory in children and young adults, olfactory identification ability was the main predictor in the elderly study group. We also calculated response bias for the separate age groups and found no differences between children, young adults and elderly. However, with age false alarm rates increased. We conclude that children possess equal olfactory sensitivity compared with adults; however, due to limitations in linguistic capabilities and familiarity to odorants, odor memory and odor identification performance was limited. Additionally, our data indicate major alterations of olfactory processing in advanced age with substantial losses in odor memory and odor identification performance.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Recent reports studying olfaction in children revealed that olfactory threshold is no different between children and young adults (Koelaga and Köster, 1974;
Perry et al., 1980; Cain et al.,1988, 1995; Dorries et al., 1989; Solbu et al., 1989). However,
olfactory identification performance was found to be poorer in children than in young adults (Dotyet al., 1985; Rothschild et al., 1995), showing
that
processes different from olfactory
sensitivity may be responsible for improved odor identification with increasing age (Richman et al., 1995).
As with other sensory modalities, olfactory capabilities are negatively correlated with age in
adults.
The elderly experience a decrease in olfactory sensitivity and a loss of suprathreshold odor
intensity
perception as measured by magnitude matching methods. Additionally, odor classification, odor
discrimination, odor identification ability and odor memory are impaired as well (Schiffman et al., 1976; Schiffman and Pasternak, 1979; Schemper et al., 1981; Doty et al.,
1984; Stevens and Cain, 1985, 1987a; Eskenazi et al., 1986; Murphy, 1986;
Murphy et
al., 1991, 1997; DeWijk and
Cain,
1994; Cain et al., 1995). The decrement of olfactory
functions has real-world significance when identification of blended food and the usefulness of
olfactory
warning agents are considered. Both identification of blended foods and detection of mercaptans
as
warning agents in cooking gas are impaired in the elderly population (Stevens et
al., 1987b;
Wysocki and Pelchat, 1993).
The underlying mechanism for the observed differences of olfactory capabilites across the
human
lifespan have not yet been untangled. Whether they are biological in nature, e.g. maturation of
cerebral
structures in children or loss of cells in the elderly, or of psychological nature, e.g. compromised
use of
cognitive operations such as active encoding, active retrieval and use of verbal labels by children
and
the elderly, have not been sufficiently examined. The relationship between odor memory and
semantic
processing in young adults has been investigated by a number of studies. In general, it has been
found
that labeling of odors enhances odor memory substantially (Engen and Ross, 1973; Rabin and Cain,
1984; Lyman and Mc Daniel, 1986, 1990; Schab, 1991; Lehrner, 1993; Schab and Crowder, 1995;
Herz and Engen, 1996).
In addition, it has been shown that labeling an odor with the same tag at studying and testing
has a
strong influence on odor recognition memory. That is, consistently labeled odors are much better
remembered than inconsistently labeled odors (Rabin and Cain, 1984; Lehrner, 1993). These data
suggest that semantic memory processes play an important role in odor memory (Larsson,
1997b).
Recent research with elderly study populations suggest that impaired odor memory is largely
attributable to cognitive factors such as odor naming. Larsson and Bäckman (1993) documented
that access to verbal labels largely determined age differences in recognition of common odors.
Additionally, Murphy et al. (1991, 1997)
reported significantly impaired odor memory and
olfactory identification performance in elderly when compared with young adults. They
suggested that
cognitive factors such as encoding, storing and retrieving appear to predict odor memory. For
children,
however, there is a lack of data regarding the importance of semantic factors in odor memory.
In recognition memory experiments, hit rates and false alarm rates are important parameters
and
are used to compute discrimination indices and response bias measures. Such measures are age
related
and they have been linked to encoding of to-be-remembered items. Studies with nonolfactory
stimuli
showed age-related increases of false alarm rates in the elderly for faces (Ferris et
al., 1980;
Bartlett and Leslie, 1986; Bartlett et al., 1989). Increased false alarm rates in the elderly have
been related more to a reliance on perceived familiarity and less on conscious recollection in
making
recognition judgements relative to the young (Bartlett et al., 1991).The issue of age-related
increases in false alarm errors has not been sufficiently investigated for olfactory stimuli. False
alarm
errors are worthwhile to investigate for the olfactory modality because considerably higher false
alarm
rates for olfactory stimuli compared with visual or auditory stimuli can be expected (Engen,
1982). Two
studies reported increased false alarm rates for the elderly compared with young adults in an
olfactory
recognition paradigm (Larsson and Bäckman, 1993; Murphy et al., 1997). However, to
our knowledge no data for children are available.
The purpose of this investigation was to assess, for the first time, odor recognition memory performance, odor identification, consistency of label use and olfactory threshold in a sample of subjects across the human lifespan. We performed this investigation in order to gather new information regarding these olfactory functions across different developmental stages. Based on the literature review, we expected a negative correlation between age and olfactory sensitivity, and an inverse U-shaped function for odor naming and memory. Another goal of this study was to address, for the first time, the relationships of olfactory sensitivity and odor naming to odor memory performance across the human life range. We hypothesized that verbal naming, i.e. semantic encoding, contributes to odor memory over the human age range, but that, specifically in elders, sensory factors such as detection threshold determines odor memory performance. In addition, we investigated whether hit rates and false alarm rates are age-related.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Subjects
A total of 142 subjects took part in the experiment. Olfactory data of 137 subjects (51 males and 86 females) were eventually used in the statistical analyses because six subjects left the experiment without completing it. Subjects were split into four age groups: children from 4 to 11 years (age group 1; n = 42), young adults from 18 to 30 years (age group 2; n = 55), middle-aged adults from 31 to 57 years (age group 3; n = 22), and elderly subjects with an age range from 64 to 90 years (age group 4; n = 18). The children were from a Vienna Elementary School and the young subjects were mostly students of the University of Vienna. Elderly people were independently living senior citizens of Vienna. For demographic characteristics of the subject groups see Table 1.
|
Method
The test battery consisted of an odor detection threshold task, an odor identification task
(everyday
odors are to be identified) and an odor recognition memory task (retention time 15 min). For the
olfactory identification task and odor memory task, 20 odorants were used. All were chosen to be
readily identifiable and are frequently used in the common household. The odorants were as
follows:
peppermint, aniseed, juicy-fruit chewing gum, turpentine, cloves, cinnamon, cocoa, coffee,
mustard,
cigarette butts, lemon, orange, shower gel, brandy, almond oil, garlic, dried coconut, soap,
gasoline,
Nivea (skin cream). A very similar testing procedure was used for assessing HIV-infected
individuals,
and patients with Alzheimer's disease and Parkinson's disease (Lehrner et al.,
1995, 1997), and the same procedural rules were
applied in the present study. The test took ~30 min.
Odor recognition memory and olfactory identification
Subjects smelled in succession 10 odorants out of the 20 item odor pool. A stimulus presentation rate of 30 s was used in order to control possible stimulus adaptation effects. While inspecting the odors, the subjects closed their eyes and usually took 1–3 sniffs lasting for ~5 s. The odorants were presented in random order for every subject in glass bottles wrapped with adhesive tape to prevent participants from visually detecting the contents. The subjects were told to name the odor as correctly as possible with one word and to memorize the odor because they would be tested later to see how well they could remember the odors. After a retention interval of 15 min the memory recognition test took place. The odorants were again given every 30 s. For the 20 stimuli, presented in random order, the subjects judged whether the odorant was `old' (i.e. a target from the inspection set) or `new'. After deciding this, the subjects named the odorant. Again they were told to use just one word. Olfactory identification ability was regarded to be a direct route to semantic knowledge of subjects about the presented odors.
Consistency of label use
The verbal encoding of the odors was analyzed in terms of consistency of label use. When a
person
uses the same verbal descriptor both at inspection and testing, very similar verbal encoding can
be
assumed. However, the use of different labels on the two occasions implies different verbal
encoding at
inspection and testing (Rabin and Cain, 1984; Lehrner, 1993). On the basis of this scoring, it was
possible to learn more about subjects' usage of semantic encoding according to their
consistency of label use. Consistency of label use was thus seen as an overt measure of the usage
of
semantic information.
Olfactory detection threshold
Odor threshold testing was established by using a 1-butanol ascending staircase, two-bottle,
forced-choice method (Cain et al., 1988). Beginning at the
lowest
concentration, subjects
received a serial dilution set of 1-butanol beginning at 4% in distilled water and progressing in 10
steps
of successive thirds (dilution factor 3) and a water control. On a given trail subjects sniffed
consecutively from two amber glass bottles and their task was to indicate which bottle contained
the
butanol solution or smelled stronger. If the subject was incorrect at one concentration, the next
higher
concentration was presented. Threshold was defined as the butanol concentration correctly
chosen
over water in four consecutive trails. The corresponding number of the concentration was taken
as the
threshold value; a high corresponding number represents a low threshold.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
First we wanted to answer the question of whether age has an influence on olfactory measures. All dependent variables were tested for differences between any of the four age groups using one-way analysis of variance (ANOVA). For significant results, Scheffe post hoc tests were used to search for reliable group differences. Figure 1 illustrates the results.
|
Odor memory
On the basis of signal detection theory a hit rate and a false alarm rate were derived. Using a
technique described by Snodgrass and Corwin (1988), the recognition
memory index Pr was
calculated. Perfect performance is rated as1.0, and 0.0 is performance at random. For the memory
measure Pr, an ANOVA yielded significant differences [F(3,133) = 19.52, P< 0.001 ]. Scheffe post hoc tests produced significant differences between
the elderly
and all other groups (elderly versus children, P< 0.001; elderly versus young adults, P< 0.001; elderly versus middle-aged adults, P< 0.001). Children had
significant
lower
scores than young adults (P< 0.02). Young adults and middle-aged adults did not
differ (P> 0.3), nor did children and middle-aged adults (P> 0.8).
Hit rates
The ANOVA on the hit rate data revealed an effect of age, [F(3,133) = 4.28, P <0.01 ]. Post hoc Scheffe tests showed that hit rate was significantly higher for young adults compared with the elderly (P< 0.03). No other differences were statistically significant (all P values >0.15).
False alarm rates
New odors incorrectly recognized as old were scored as false alarms. The ANOVA on the number of false alarms yielded an effect of age [F(3,133) = 20.46, P< 0.001 ]. Scheffe post hoc testing revealed that the elderly produced a higher false alarm rate than all other groups (elderly versus children, P< 0.001; elderly versus young adults, P< 0.001; elderly versus middle-aged adults, P< 0.001). No other differences were significant (all P values >0.15).
Response criterion
For the measure of the response criterion we used the bias index (Br) suggested by
Snodgrass and Corwin (1988). Accordingly, we corrected hit and false
alarm
rates by adding 0.5 to
each frequency and divided by n + 1, where n is the number of old or new
trials. For
the computation of Br the transformed values were used. A value of Br of 0.5
indicates neutral bias, a value >0.5 indicates liberal bias and a value <0.5 indicates
conservative
bias.
The ANOVA on Br data revealed a significant age effect [F(3,133) = 3.60, P< 0.02 ]. Post hoc Scheffe tests showed that the elderly had a significantly
higher Br score than middle-aged adults (P< 0.02). No other pairwise
comparisons were
significant (all P values >0.2).
Olfactory identification
Odor identification scores were derived from the naming of the 20 odors at odor memory
testing.
The word given for identification was scored for correctness. The response to each stimulus was
categorized into three levels of correctness using the scheme described by Cain (1979). He categorized
the generated odor labels of his subjects into three groups: a veridical label (the true name of the
odorant, e.g. gasoline for gasoline; coffee for coffee); a near miss, i.e. names reasonably close to
the
veridical name (e.g. cinnamon for cloves); and a far miss, i.e. names quite far from veridical
labels (e.g.
lemon for coffee). Veridical name, near miss and far miss were designated scores of 2, 1 and 0
respectively. A subject could score at most 40 points for the identification of the 20 odors.
Olfactory identification showed a significant age effect [F(3,133) = 37.8 P< 0.001 ] and single Scheffe post hoc comparisons were significant between young adults and all other groups (young adults versus children, P< 0.001; young adults versus middle-aged adults, P< 0.01; young adults versus elderly, P< 0.001). Additionally, children and middle-aged adults scored significantly better than the elderly (children versus elderly, P< 0.001; middle-aged adults versus elderly, P< 0.007). Children and middle aged adults did not differ (P> 0.08).
Consistency of label use
Subjects were scored on how consistently they used their generated odor label. When a
person
labeled an odor with the same tag at presentation and testing (e.g. coffee–coffee), same
association (SA) was scored. Different association (DA) was scored when a subject labeled an
odor
differently at initial presentation and testing (e.g. coffee–chocolate). No association (NA)
was
scored when a subject was not able to generate a label for the odor at presentation and testing, or
when a subject could generate only one label either at presentation or at testing. DA and NA
scores
were pooled. This is a modified version of our previously used categorization method (Lehrner, 1993).
The number of consistently labeled odors had an overall age effect [F(3,133) = 7.5, P< 0.001 ]. Post hoc Scheffe analyses showed significant differences between
young
adults and children (P< 0.003), and between young adults and the elderly (P<
0.003). Further pairwise comparisons revealed that young adults versus the middleaged adults,
middle-aged adults versus children and children versus the elderly all did not differ (all P values
>0.30).
Olfactory threshold
The comparison of olfactory threshold across age groups did not reveal a statistical difference. However, a trend was present [F(3,127) = 2.5, P< 0.06 ].
Scheffe pairwise comparisons showed a trend only for children versus elderly (P< 0.06). All other comparisons were not significant (all P values >0.16).
Within-group variability
In order to investigate whether there were differences in olfactory performance within the group of children we calculated the age median of children (median = 9 years) and then split the sample into two halves (young children: n = 20, mean age 6.8 ± 1.2; older children: n = 21, mean age = 9.6 ± 0.7 years). Subsequently we performed separate Mann–Whitney U-tests for the variables of odor memory, hit rate, false alarm rate, response criterion, olfactory identification, consistency of label use and olfactory threshold for the two independent samples. Statistical analysis revealed no significant effect of age group on any variable (all P values >0.3).
In addition, the elderly population was split along the age median (median = 78) with corresponding two groups (young elderly: n = 9, mean age 70.8 years and older elderly: n = 9, mean age 84.8 years, respectively). Subsequent separate Mann–Whitney U-tests for the variables of odor memory, hit rate, false alarm rate, response criterion, olfactory identification, consistency of label use and olfactory threshold were performed. Mann–Whitney U-tests revealed significant differences for the variables of hit rate (U = 18.5; P< 0.05), false alarm rate (U = 12.5; P< 0.02), response criterion (U = 13.0; P< 0.02) and olfactory threshold (U = 11.0; P< 0.03), whereas odor memory, consistency of label use and olfactory identification showed no significant differences between young elderly and old elderly (Mann–Whitney U-tests; all P values >0.10). Table 2 shows the results.
|
Analysis of predictors of odor memory performance
One special interest of this study was to investigate the potential influence of verbal identification, consistency of label use and olfactory threshold on odor memory across the human lifespan. This was done to untangle sensory influences, e.g. olfactory threshold, from more cognitively influenced measures such as odor identification and consistency of label use. In order to investigate this issue, we calculated analyses of covariance (ANCOVA) with olfactory threshold, odor identification and consistency of label use as covariates. ANCOVA compares group means after adjusting those means for a third variable, or covariate. If differences in group means could be explained based on the relationship with the third variable, results from the ANCOVA would not be statistically significant. Controlling for olfactory threshold, odor identification and consistency of label use with separate ANCOVAs did not change results. When adjusting for all three covariates, significant age differences for Pr were still apparent [F(3,124) = 4.0, P< 0.009 ].
The ANCOVA does not directly evaluate the strength of association between two variables and is thus not helpful in elucidating the relationship between odor memory and the variables of olfactory threshold, odor identification and consistency of label use. Because of the non-monotonic changes in performance for odor memory, odor identification and consistency of label use across age, we calculated product–moment correlations for the adults groups only, between chronological age, olfactory threshold, odor identification, consistency of label use and odor memory, and we also calculated correlation coefficients for the whole sample. The correlational analyses for adults indicated a significant negative relationship between age and odor identification, odor memory, consistency of label use and olfactory threshold. Olfactory threshold correlated negatively with both odor memory and odor identification. Highly significant positive relationships were found between odor identification, consistency of label use and odor memory performance. Table 3A shows the results. An interesting finding is the significant negative correlation between false alarm rate and olfactory threshold (r = -0.37). A closer inspection of the correlations between false alarm rate and threshold in each age group revealed nonsignificant correlations in children (r = -0.05), young adults (r = -0.23) and middle-aged adults (r = -0.18) respectively. However, a significant negative correlation emerged in the elderly age group (r = -0.53; P< 0.05).
|
When including children in the analysis, the pattern of correlations among performance scores remained the same although the correlation coefficients usually decreased slightly. Table 3B shows the results.
In order to analyze the interrelationships between olfactory threshold, odor identification, consistency of label use and odor memory, multiple regression analyses for the whole sample and for each age group separately were calculated. Olfactory threshold, odor identification and consistency of label use were used as predictor variables and Pr was the dependent or criterion measure. Data were analyzed by stepwise regression analysis.
Table 4 shows the results of the regression analyses including the standardized beta weights for the significant predictor variables. First we report our results concerning the whole sample. Regression analysis was highly significant (P< 0.001), accounting for 42% of the variance. Two variables were powerful predictors, with olfactory identification being highly associated and consistency of label use being moderately associated to odor memory. Olfactory threshold was not a significant predictor and thus was removed from the equation.
|
Separate multiple regression analyses revealed that in the subsamples of children, young adults and middle-aged adults consistency of label use was the main predictor of odor memory performance. Stepwise regression analysis for the elderly group revealed odor identification as the main predictor. Consistency of label use and olfactory threshold did not reach the significance level of
=
0.05 and
thus were removed. However, the partial correlation coefficient between odor memory and
olfactory
threshold (r part = 0.47, P> 0.08) indicated a statistical trend. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The present experiments sought to clarify some questions regarding olfactory functions during the human life cycle and the relationship of odor detection sensitivity and odor naming to odor memory performance. Our investigation confirmed that olfactory sensitivity is well developed in children although odor recognition memory performance and odor naming are somewhat limited. Additionally, we confirmed a marked decline in olfactory functioning for the elderly, which is indicative of alterations in the olfactory system. Semantic encoding is correlated with odor memory to a varying degree over the human lifespan. Whereas in children and young adults, semantic context is the main predictor of odor memory, odor identification ability is the main predictor in the elderly. Deficits for both hits and false alarms in odor recognition were also found in the elderly. Motivational factors, as expressed by response bias parameters, were not different between children, young adults and the elderly study group. Thus, we found converging evidence suggesting major changes in processing of olfactory stimuli across the human age range.
We confirmed that children are as sensitive as young adults for 1-butanol. As age increases,
olfactory sensitivity decreases in a monotonic fashion. This is indicated by a negative correlation
coefficient and a trend in the ANOVA. Our data are in accordance with previous reports in the
literature (Koelega and Köster, 1974; Dorries et al., 1989; Cain et al., 1995).
For the more cognitively demanding tasks of odor memory, consistency of label use and odor
identification we obtained clear-cut results. Odor recognition memory, consistency of label use
and
olfactory identification showed a non-monotonic change with increases from childhood to
adulthood
and then a decrease throughout adulthood with impaired functioning in the elderly. Our findings
confirm
previous reports for olfactory identification (Cain et al., 1995)
and are
novel for odor memory
and consistency of label use.
The functional importance of the sense of smell in children has been demonstrated by other
investigators as well (Engen, 1986; Schmidt and Beachamp,
1988; Schaal, 1988; Porter, 1991).
Regarding the human sense of smell in infants, Sullivan et al. (1991)
provided the first evidence
that infants are capable of olfactory associative learning during the first day after birth. This
finding is
remarkable because it shows that the infant odor memory system is independent of semantic
memory
functioning.
Slow acquisition of names for odors has been reported in adults (Davis, 1975, 1977). Recently,
Cain et al. (1995) provided evidence that children improved
faster
than adults on explicit
paired-associate learning of names for odors. However, they started from an initially lower level
of
semantic knowledge of test odors.
Engen and Engen (1997) reviewed language development studies in
naturalistic settings and found
that (i) research of the development of the lexicon revealed that children are more concerned with
color, size and location of objects than with smells and (ii) that there are virtually no examples of
children spontaneously using the perception verb `smell' to respond to odor
experiences.
Although acquisition of verbal labels may be incidental and may thus proceed slowly during child development in a natural setting, children in the age range of 4-11 years have already acquired a substantial odor vocabulary. They are capable of using this semantic knowledge for encoding olfactory information and thus semantic encoding of odors is important for odor memory processes during early stages of childhood. This was clearly indicated by stepwise multiple regression analysis showing that consistent verbal labeling of odors is a good predictor of odor memory performance in children. Our data show that, once a child has named an odor consistently, he or she will remember this odor very well.
In summary, we confirmed that odor naming and odor memory were less developed in children. Two hypotheses are put forward to explain this phenomenon. First, a reasonable explanation could be that children may not yet have learned some of the verbal descriptors, and thus a certain odor is not identified correctly because the odor-related vocabulary is missing in semantic memory. Second, children have not had the opportunity to smell all the presented odors previously, and consequently the odors are unfamiliar to them. Thus, no memory trace in semantic memory is available. Whatever the reason, it is important to note that olfactory sensitivity is not impaired in children, and thus cannot account for poor odor naming and odor memory performance. Thus, we suggest that children have the capacity to detect odors in a comparable fashion to adults. However, naming of odorants slowly develops from childhood to adulthood and as a consequence episodic odor recognition memory of children is inferior than that of adults.
The elderly population in our study also showed clear-cut results. Performance was markedly reduced in the elderly for odor identification, odor memory and consistency of label use. For the measure of olfactory threshold we observed no statistical difference. However, compared with children a trend was detected (P< 0.06), indicating that the elderly also experienced a loss of olfactory sensitivity.
The mechanism of the olfactory loss in the elderly is not well understood. There is evidence
that
physiological and neuropathological changes in the olfactory system, such as age-related changes
in the
nasal cavity, and alterations in the olfactory mucosa and the olfactory bulb, are a common feature
of
growing older (Doty, 1991). On the other hand, impaired cognitive
abilities
could also be a major
factor contributing to the decreased performance of the elderly in olfactory behavioral tests. For
example, it is clear that cognitive capabilities decrease in the elderly, especially memory
functions. Some
cognitive processes involve more effortful processing than others (e.g. recall versus recognition),
thus
older people are relatively penalized when such processing is required (Craik, 1983). Further studies of
olfactory functions in elderly populations should include naming and memory tests in other
modalities in
order to establish the relationship between olfaction and cognition in the elderly.
An interesting question concerns the turning point of olfactory functions in children and elderly. As our analysis demonstrated, there was no difference for olfactory threshold, odor memory and odor identification between children aged 4–8 years and aged 9–11 years, thus olfactory functions may develop at some point earlier than the age groups tested. Young adults are best at olfactory sensitivity, odor identification and odor memory; middle-aged adults experience a gradual decline in their olfactory capabilities until the age of >60 years, when there is a clear loss of olfactory identification ability and odor memory. The decrement of olfactory threshold is not as dramatic. However, elderly subjects aged >78 years also show considerable odor detection impairment.
As demonstrated in recent studies (Larsson and Bäckman, 1993; Larsson, 1997a), we also
found that odor naming and odor memory were strongly related. However, our analysis indicated
consistency of label use to be the main predictor of odor memory instead of odor identification
ability.
This is a new finding. Thus, naming a particular odor, both at learning and testing with the same
label,
independent of the ecological correctness of this label, determines odor memory performance for
that
odor in children, young adults and middle-aged adults, i.e. the personally generated label is more
important for odor recognition than the ecological name of the odor. As has been pointed out by
Engen
(1982), the link between odor perception and language may be weak.
Odor
names are somewhat
arbitrary and artificial. However, odor identification and consistency of label use are interrelated,
indicating that a rich semantic trace of an odor sensation enhances consistent odor naming and
subsequently odor recognition.
Our data suggest that there is a shift of olfactory processing in the elderly. Whereas in children and young adults, consistency of label use—that is, verbal naming context—is the main predictor of odor memory, in the elderly odor identification ability becomes the main predictor. However, the detection threshold was marginally significantly correlated with odor memory, indicating that primary sensory processes might be responsible for the poor odor memory in elderly subjects. It appears that children and young adults are able to successfully use consistent labeling—that is, generating the same verbal tag both at learning and testing—as a memory aid. In contrast, due to their slightly elevated threshold, the elderly no longer have the ability to discriminate odors as accurately as the young, and thus odors are no longer consistently named. This considerably alters their ability to remember odors because semantic processes are an important part of odor memory. As a consequence, the elderly remember only familiar odors they can also correctly identify.
An interpretation of the results of multiple stepwise regression analysis for the whole sample
indicates an association between odor identification and consistency of label use—that is,
`semantic encoding' and odor memory. However, ANCOVA results show that
large
parts of the variance of odor memory cannot be explained by olfactory threshold, consistency of
label
use or odor identification alone, indicating that other factors must also be involved in odor
memory
processes. We suggest that familiarity judgements on the basis of perceptual aptitude are
important
determinants of odor memory, as has been suggested by dual-process models of recognition
memory
for other stimuli as well (Mandler, 1980).
To study dual-process memory the olfactory modality may be particularly rewarding because
odors produce a prominent tip-of-the-nose phenomenon (Lawless and Engen, 1977) and unfamiliar
odors may be remembered quite well (Engen and Ross, 1973). For
instance,
recent evidence showed
that odors can be remembered quite well without adequate verbal labeling. Earlier reports
documented
that giving an odor at presentation and testing the same verbal tag results in a recognition
accuracy of
87% (Lehrner, 1993). However, giving an odor different labels at
presentation
and testing is indicative
of somewhat different semantic encoding, and still produces recognition rates above chance, with
~70% correct recognition (Lehrner, 1993). Such processing of
olfactory
stimuli may be similar to the
experimental approach of the `remember' and `know' paradigm (Tulving,
1985). According to this distinction, recognition measured by a
`remember'
response
indicates that recognizing the test item brings to mind some conscious recollection of its prior
occurrence in the learning list, such as an association it has triggered. A `know'
response
is assumed to indicate that recognizing the item brings to mind a feeling of familiarity. That is,
the subject
is able to tell that a given item has previously been presented, but does not have a recollection of
its
prior occurrence in the study list (Mäntlya, 1993).
Remember/know
experiments showed
consistently better recognition for the remember items. However, `know' items
were
remembered above chance level as well (Rajaram, 1993).
Recent research suggested that the increase in false alarm rate in the elderly may be related to
the
fact that they rely less on conscious recollection in recognition experiments than on perceived
familiarity,
that is, the feeling of having had an item before but no conscious recollection of that particular
item
(Bartlett et al., 1991). False alarm rate and hit rate analysis
showed
straightforward results for
the elderly. They had higher false alarm rates and lower hit rates than young adults. This finding
is
consistent with several previous studies for olfactory stimuli (Larsson and
Bäckman,
1993;
Murphy et al., 1997) and for faces and songs (Bartlett et al., 1991; 1989; 1995).
The
response criterion did not differ between young adults and the elderly, indicating no difference in
response behavior for the two study groups. However, response criterion did differ between
younger
and older elderly (Table 2), indicating a more liberal response criterion
for the
older elderly. We also
found a negative correlation between olfactory threshold and false alarm rate in the elderly,
showing that
with decreased olfactory sensitivity the elderly may not discriminate the odors properly and thus
tend to
call a new odor old. Thus, our data suggest that recognition judgements in the elderly may be
based to
a large extent on perceived familiarity and such perceptually driven processing is dependent on
olfactory sensitivity. In contrast to the elderly, there is no significant correlation between
olfactory
threshold and false alarm rate in children, young adults and middle-aged adults. Hit rate, false
alarm rate
and response criterion did not differ between children and young adults, indicating that basically
the
same motivational factors regarding olfaction are at work in the three study groups. The finding
that
false alarm rate in children is not significantly different from young adults indicates that children
may rely
more on conscious recollection. However, their ability to label odors consistently is significantly
poorer
compared with young adults. Thus, children may be in between young adults and the elderly in
using
perceived familiarity for recognition of odors.
Our data suggest that recognition of odors may involve explicit, context-dependent memory
processes and implicit, data-driven memory processes. The investigation of olfactory perceptual
fluency
for odor memory processes and its incorporation into modern memory concepts of explicit/
implicit
memory (Schacter, 1990; DeSchepper and Treisman, 1996; Snodgrass et al., 1996) seems
worthwhile for the future.
|
Acknowledgments |
|---|
We would like to thank Ms Sylvia Philipp for her generous cooperation in the conduct of this study and Prof. Trygg Engen for providing the initial stimulating ideas for this experiment. We would also like to thank Ms Maggie Lee Huckabee for making helpful comments on earlier drafts of this paper.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Bartlett, J.C. and Leslie, J.E. (1986) Aging and memory for faces versus single views of faces. Mem. Cogn., 14, 371–381.[Web of Science][Medline]
Bartlett, J.C., Leslie, J.E., Tubbs, A.and Fulton, A. (1989) Aging and memory for pictures of faces. Psychol. Aging, 4,276 –283.[Web of Science][Medline]
Bartlett, J.C., Strater, L. and Fulton, A. (1991) False recency and false fame of faces in young adulthood and old age. Mem. Cogn., 19,177 –188.[Web of Science][Medline]
Bartlett, J.C., Halpern, A.R. and Dowling, W.J. (1995) Recognition of familiar and unfamiliar melodies in normal aging and Alzheimer`s disease. Mem. Cogn., 23, 531–546.
Cain, W.S. (1979) To know with the nose: keys to odor identification. Science, 203, 467–470.
Cain, W.S., Gent, J.F., Goodspeed, R.B. and Leonard, G. (1988) Evaluation of olfactory dysfunction in the Connecticut Chemosensory Clinical Research Center.Laryngoscope , 98, 83–88.[Web of Science][Medline]
Cain, W.S., Stevens, J.C., Nickou, C.M., Giles, A., Johnston, I. and Garcia-Medina, M.R. (1995) Life-span development of odor identification, learning, and olfactory sensitivity. Perception, 24, 1457–1472.[Web of Science][Medline]
Craik, F.I.M. (1983) On the transfer of information from temporary to permanent memory. Phil. Trans. Roy. Soc. Lond., 302,341 –359.
Davis, R.G. (1975) Acquisition of verbal associations to olfactory stimuli of varying familiarity and to abstract visual stimuli. J. Exp. Psychol.: Hum. Learn. Mem., 104, 143–142.
Davis, R.G. (1977) Acquisition and retention of verbal associations to olfactory and abstract visual stimuli of varying similarity. J. Exp. Psychol.: Hum. Learn. Mem., 3, 37–51.
DeSchepper, B. and Treisman, A. (1996) Visual memory for novel shapes: implicit coding without attention. J. Exp. Psychol: Learn. Mem. Cogn., 22, 27–47.[Web of Science][Medline]
DeWijk, R. and Cain, W.S. (1994) Odor identification by name and by edibility: life-span development and safety. Hum. Factors, 36,182 –187.[Web of Science][Medline]
Dorries, K.M., Schmidt, H.J., Beauchamp, G.K. and Wysocki, C.J. (1989) Changes in sensitivity to the odor of androstenone during adolescence. Devl Psychobiol., 22, 423–435.
Doty, R.L. (1991) Influences of aging on human olfactory function. In Laing, D.G. Doty, R.L. and Breipohl, W. (eds), The Human Sense of Smell. Springer-Verlag, Berlin, pp. 155-163.
Doty, R.L., Shaman, P., Applebaum, S.L., Giberson, R., Sikorski, L.
and
Rosenberg, L. (1984) Smell identification ability: changes with age. Science, 226, 1441–1443.
Doty, R.L. Applebau, S. Zusho, H. and Settle, R.G. (1985) Sex differences in odor identification ability: a cross cultural analysis. Neuropsychologia, 23, 667–672.[Web of Science][Medline]
Engen, T. (1982) The Perception of Odors. Academic Press, Toronto.
Engen, T. (1986) Children's sense of smell. In Meiselman, H.L. and Rivlin, R.S. (eds), Clinical Measurement of Taste and Smell. Macmillan, New York, pp. 316-317.
Engen, T. and Engen E. (1997) Relationship between development of odor perception and language. Enfance, 1, 125–140.
Engen, T. and Ross, B.M. (1973) Long-term memory of odors with and without verbal descriptions. J. Exp. Psychol., 100, 221–227.[Web of Science][Medline]
Eskenazi, B., Cain, W.S. and Friend, K. (1986) Exploration of olfactory aptitude. Bull. Psychon. Soc., 24, 203–206.[Web of Science]
Ferris, S.H., Crook, T., McCarthy, M. and Rae, D. (1980) Facial
recognition memory deficits in normal aging and senile dementia. J. Gerontol., 35, 707–714.
Herz, R.S. and Engen, T. (1996) Odor memory: review and analysis. Psychon. Bull. Rev., 3, 300–313.[Web of Science]
Koelega, H.S. and Köster E.P. (1974) Some experiments on sex differences in odor perception. Ann. N.Y. Acad. Sci., 237,234 –246.[Web of Science][Medline]
Larsson, M. (1997a) Age-related differences in episodic odor recognition: the role of access to specific odor names. Memory, 5, 361–378.[Web of Science][Medline]
Larsson, M. (1997b) Semantic factors in episodic
recognition of
common odors in
early and late adulthood: a review. Chem. Senses, 22, 623–633.
Larsson, M. and Bäckman, L. (1993) Semantic activation and episodic odor recognition in young and older adults. Psychol. Aging, 8,582 –588.[Web of Science][Medline]
Lawless, H. and Engen, T. (1977) Associations to odors: interference, mnemomics and verbal labeling. J. Exp. Psychol.: Hum. Learn. Mem., 3, 52–59.
Lehrner, J. (1993) Gender differences in long-term
odor
recognition
memory: verbal versus sensory influences and consistency of label use. Chem. Senses, 18, 17–26.
Lehrner, J.P., Kryspin-Exner, I. and Vetter, N. (1995a) Higher olfactory threshold and decreased identification ability in HIV-infected persons. Chem. Senses, 20, 325–328.
Lehrner, J.P., Brücke, T., Dal-Bianco, P., Gatterer, G. and
KryspinExner, I. (1997) Olfactory functions in Parkinson's
disease and
Alzheimer's
disease. Chem. Senses, 22, 105–110.
Lyman, B.J. and McDaniel, M.A. (1986) Effects of encoding strategy on long-term memory for odors. Quart. J. Exp. Psychol., 38A,753 –765.
Lyman, B.J. and McDaniel, M.A. (1990) Memory for odors and odor names: modalities of elaboration and imagery. J. Exp. Psychol.: Learn. Mem. Cogn., 16, 656–664.[Web of Science]
Mandler, G. (1980) Recognizing: the judgement of previous occurrence. Psychol. Rev., 87, 252–271.
Mäntyla, T. (1993) Knowing but not remembering: adult age differences in recollective experience. Mem. Cogn., 21, 379–388.[Web of Science][Medline]
Murphy, C. (1986) Taste and smell in the elderly. In Meiselman, H.L. and Rivlin, R.S. (eds), Clinical Measurement of Taste and Smell. Macmillan, New York, pp. 343-371.
Murphy, C., Cain, W.S., Gilmore, M.M. and Skinner, B. (1991) Sensory and semantic factors in recognition memory or odors and graphic stimuli: elderly versus young persons. Am. J. Psychol., 104, 161–182.[Web of Science][Medline]
Murphy, C., Nordin, S. and Acosta, L. (1997) Odor learning, recall and recognition memory in young and elderly adults. Neuropsychology, 11,126 –137.[Web of Science][Medline]
Perry, J.D., Frisch, S., Jafek, B. and Jafek, M. (1980) Olfactory detection thresholds using pyridine, thiophene, and phenylethyl alcohol. Otol. Head Neck Surg., 88, 778–782.
Porter, R.H. (1991) Human reproduction and the mother-infant relationship: the role of odors. In Getchell, T.V., Doty, R.L., Bartoshuk, L.M. and Snow, J.B. (eds), Smell and Taste in Helath and Disease. Raven Press, New York, pp. 429-442.
Rabin, M.D. and Cain, W.S, (1984) Odor recognition: familiarity, identifiability, and encoding consistency. J. Exp. Psychol: Learn. Mem. Cogn.,10 , 316–325.[Web of Science][Medline]
Rajaram, S. (1993) Remembering and knowing: two means of access to the personal past. Mem. Cogn., 21, 89–102.[Web of Science][Medline]
Richman, R.A. Wallace, K. and Sheehe, P.R. (1995) Assessment of an abbreviated odorant identification task for screening: a rapid screening device for schools and clinics.Acta Paed. , 84, 434–437.[Web of Science][Medline]
Rothschild, M.A., Myer, C.M. and Duncan, H.J. (1995) Olfactory disturbance in pediatric tracheotomy. Otol. Head Neck Surg., 111,71 –76.
Schaal, B. (1988) Olfaction in infants and children:
developmental
and functional perspectives. Chem. Senses, 13, 145–190.
Schab, F.R. (1991) Odor memory: taking stock. Psychol. Bull., 2, 242–251.
Schab, F.R. and Crowder, R.G. (1995) Memory for Odors. Lawrence Erlbaum Associates, Hilldale, NJ.
Schacter, D.L. (1990) Perceptual representation systems and implicit memory: toward a resolution of multiple memory systems debate. Ann. N.Y. Acad. Sci., 608, 543–571.[Web of Science][Medline]
Schemper, T.S., Voss, S. and Cain, W.S. (1981) Odor
identifiation
in young and elderly persons. Sensory and cognitive limitations. J. Gerontol., 36, 446–42.
Schiffman, S. and Pasternak M. (1979) Decreased discrimination of food odors in the elderly. J. Gerontol., 34, 73–79.[Abstract]
Schiffman, S.S., Moss, J. and Erickson, R.P. (1976) Thresholds of food odors in the elderly. Exp. Aging Res., 2, 389–398.[Medline]
Schmidt, H.J. and Beauchamp, G.K. (1988) Adult-like odor preferences and aversions in three-year-old children. Child Dev., 59,1136 –1143.[Web of Science][Medline]
Snodgrass, J.G. and Corwin, J. (1988) Pragmatics of measuring recognition memory: applications to dementia and amnesia. J. Exp. Psychol.: Gen., 117, 34–50.[Web of Science][Medline]
Snodgrass, J.G., Hirshman, E. and Fan, J. (1996) The sensory match effect in recognition memory. Perceptual fluency or episodic trace? Mem. Cogn., 24, 367–383.
Solbu, E.H., Jellestad F.K. and Straetkvern, K.O. (1989) Children's sensitivity to odor of trimethylamine. J. Chem. Ecol., 16,1829 –1840.[Web of Science]
Stevens, J.C. and Cain, W.S. (1985) Age-related deficiency
in
the
perceived strength of six odorants. Chem. Senses, 10, 517–529.
Stevens, J.C. and Cain, W.S. (1987a) Old-age deficits in the sense of smell as gauged by thresholds, magnitude matching and odor identification. Psychol. Aging, 2, 36–42.[Web of Science][Medline]
Stevens, J.C., Cain, W.S. and Weinstein, D.E. (1987b) Aging impairs the ability to detect gas odor. Fire Technol., 23, 198–204.
Sullivan, R., Taborsky, S., Mendoza, R., Itano, A., Leon, M., Cotman, C.,
Payne, T.
and Lott, I. (1991) Olfactory classical conditioning in neonates. Pediatrics, 87, 511–518.
Tulving, E. (1985) Memory and consciousness. Can. Psychol., 26, 1–12.
Wysocki, C.J. and Pelchat, M.L. (1993) The effects of aging on the human sense of smell and its relationship to food choice. Crit. Rev. Food Sci. Nutr., 33, 63–82.[Web of Science][Medline]
Accepted February 17, 1999
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. C. Lee, H. Tian, X. Grosmaitre, and M. Ma Expression Patterns of Odorant Receptors and Response Properties of Olfactory Sensory Neurons in Aged Mice Chem Senses, October 1, 2009; 34(8): 695 - 703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ferdenzi, G. Coureaud, V. Camos, and B. Schaal Human awareness and uses of odor cues in everyday life: Results from a questionnaire study in children International Journal of Behavioral Development, September 1, 2008; 32(5): 422 - 431. [Abstract] [PDF]
|
|
|
|
 |
|
 A. FUSARI and S. BALLESTEROS Identification of odors of edible and nonedible stimuli as affected by age and gender Behav Res Methods, August 1, 2008; 40(3): 752 - 759. [Abstract] [PDF]
|
|
|
|
|
|
M. Bensafi, F. Rinck, B. Schaal, and C. Rouby Verbal Cues Modulate Hedonic Perception of Odors in 5-Year-Old Children as well as in Adults Chem Senses, November 1, 2007; 32(9): 855 - 862. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mackay-Sim, A. N. B. Johnston, C. Owen, and T. H. J. Burne Olfactory Ability in the Healthy Population: Reassessing Presbyosmia Chem Senses, October 1, 2006; 31(8): 763 - 771. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Hudson, A. Arriola, M. Martinez-Gomez, and H. Distel Effect of Air Pollution on Olfactory Function in Residents of Mexico City Chem Senses, January 1, 2006; 31(1): 79 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Sun and B. P. Halpern Identification of Air Phase Retronasal and Orthonasal Odorant Pairs Chem Senses, October 1, 2005; 30(8): 693 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Lumeng, M. D. Zuckerman, T. Cardinal, and N. Kaciroti The Association between Flavor Labeling and Flavor Recall Ability in Children Chem Senses, September 1, 2005; 30(7): 565 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Sulmont-Rosse, S. Issanchou, and E.P. Koster Odor Naming Methodology: Correct Identification with Multiple-choice versus Repeatable Identification in a Free Task Chem Senses, January 1, 2005; 30(1): 23 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. I Rupp, W. W. Fleischhacker, G. Kemmler, H. Oberbauer, A. W Scholtz, C. Wanko, and H. Hinterhuber Various Bilateral Olfactory Deficits in Male Patients With Schizophrenia Schizophr Bull, January 1, 2005; 31(1): 155 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E.P. Koster Does Olfactory Memory Depend on Remembering Odors? Chem Senses, January 1, 2005; 30(suppl_1): i236 - i237. [Full Text] [PDF]
|
|
|
|
|
|
M. Larsson, L.-G. Nilsson, J. K. Olofsson, and S. Nordin Demographic and Cognitive Predictors of Cued Odor Identification: Evidence from a Population-based Study Chem Senses, July 1, 2004; 29(6): 547 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. V. Parr, D. Heatherbell, and K. G. White Demystifying Wine Expertise: Olfactory Threshold, Perceptual Skill and Semantic Memory in Expert and Novice Wine Judges Chem Senses, October 1, 2002; 27(8): 747 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Laska Perception of Trigeminal Chemosensory Qualities in the Elderly Chem Senses, July 1, 2001; 26(6): 681 - 689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Larsson, D. Finkel, and N. L. Pedersen Odor Identification: Influences of Age, Gender, Cognition, and Personality J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2000; 55(5): 304P - 310. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||