ABSTRACT
Objective
Are paired chemosensory stimuli with very short interstimulus‐intervals (ISIs) more often perceived as distinct events when delivered synchronously with inspiration than when delivered asynchronously?
Methods
Twin stimuli (H2S, phenylethyl alcohol [PEA], CO2) were repeatedly administered intranasally to 40 participants using a constant‐flow olfactometer, synchronized with inspiration, at varying ISIs. In the first part of the study, participants indicated whether they perceived one or two stimuli. In the second part, event‐related potentials (CSERP) were recorded from 20 participants using only H2S stimuli with very short ISIs (0.6–1.6 s). The CSERPs were then qualitatively and quantitatively evaluated based on whether stimuli were perceived or not.
Results
The perception rate of twin stimuli increased with increasing ISI. A plateau in the discrimination of separate twin stimuli, averaging 90%, was observed with CO2 or H2S stimulation at ISIs between 1.6 and 3.0 s, respectively, up to 9.0 s. For PEA stimulation, a plateau of approximately 70% was observed between 3.0 and 9.0 s. CSERP detection rates were 100% for both perceived first and second twin stimuli, while non‐perceived second stimuli showed a 70% rate. The latencies of the CSERPs did not differ across the three conditions. The amplitudes were larger for non‐perceived second stimuli than for perceived first and second stimuli.
Conclusion
Predominantly inspiration‐synchronous stimulation allows higher detection of responses at both psychophysical and electrophysiological levels, even at very short ISIs. These findings may help to reduce the duration of clinical experiments and electrophysiological assessments of olfactory function.
Level of Evidence
4.
Keywords: discrimination, olfaction; flow‐olfactometer, olfactory‐evoked potentials, twin‐stimuli
Predominantly inspiration‐synchronous stimulation allows higher detection of responses at both psychophysical and electrophysiological levels, even at very short ISIs. These findings may help to reduce the duration of clinical experiments and electrophysiological assessments of olfactory function.
1. Introduction
This study builds on prior research investigating the temporal processing of repeated, identical chemosensory stimuli. The reference study by Schriever et al. investigated the minimum interstimulus interval (ISI) required for the separate perception of two identical olfactory or trigeminal stimuli in healthy subjects [1]. For this purpose, participants received suprathreshold stimulus pairs of H2S, phenylethyl alcohol (PEA) or CO2, intranasally via a computer‐controlled constant‐flow olfactometer, with short ISIs ranging from 2 to 9 s. Stimuli were administered asynchronously during mouth breathing. This study demonstrated that the likelihood of perceiving stimulus pairs as distinct increased with longer ISI. Another focus of Schriever et al.'s work was the central nervous system processing of olfactory stimuli [1].
The finding that human's ability to perceive identical olfactory stimuli separately declines at shorter ISIs has been corroborated by other studies [2, 3, 4]. In all these studies, stimulation occurred independently of the respiratory cycle. However, additional evidence suggests that the inspiration phase plays a role in modulating olfactory information processing [5, 6, 7, 8, 9, 10]. Consequently, in the present study, stimuli were automatically triggered synchronously with inspiration.
The first part of this study aimed to test the hypothesis that, in comparison to the reference study, inspiration‐synchronous stimulus application would improve the discrimination rate of separated twin stimuli with different ISIs and that the discrimination rate depends on stimulus quality. We used the same flow‐olfactometer type, the same stimuli qualities (CO2 as trigeminal and H2S as unpleasant and PEA as pleasant olfactory stimuli), and similar ISIs, as in the Schriever et al. study. The second part of the experiment was to test the hypothesis that the overall rate of detected olfactory chemosensory event‐related potentials (CSERPs) would change under inspiration‐synchronous stimulation and how the parameters of the CSERP behave with perceived and non‐perceived stimuli.
2. Materials and Methods
The study protocol was approved by the Ethics Committee of Friedrich Schiller University Jena (reference number: 2018–1177‐BO).
2.1. Inclusion and Exclusion Criteria
Inclusion criteria: Age ≥ 18 years, an identification score > 12 on the Sniffin' Sticks 16‐item odor identification test (Burghart Messtechnik, Holm, Germany), healthy subjects without known pre‐existing conditions; non‐smoking for at least 2 h prior to testing.
Exclusion criteria: Acute and chronic diseases of the upper respiratory tract, pre‐existing neurological or psychiatric conditions, use of medications known to affect the sense of smell.
The study comprised two parts (Table 1): The first part investigated the psychophysical discrimination of olfactory twin stimuli, and the second part assessed the electrophysiological discrimination using CSERP. In both parts, the first of the twin stimuli was triggered synchronously with inspiration. The second of the twin stimuli was time‐triggered by ISI. Forty healthy adults were included in the first part (25 women, 15 men; mean age: 23.6 ± 2.8 years; median: 23.0 years). Twenty participants from the first part were included in the second part, who perceived the H2S twin stimuli as a single stimulus in approximately 50% of cases during very short ISIs (ISI 600 ms: n = 6, ISI 1200 ms: n = 13, ISI 1600 ms: n = 1). This subgroup study population comprised 11 women and 9 men (mean age: 23.9 ± 2.8 years; median: 23.5 years).
TABLE 1.
Overview of the experimental procedures. ISI: Interstimulus interval; ITI: Intertrial interval; PEA: Phenylethyl alcohol.
| Part 1 of the study | Part 2 of the study | |
|---|---|---|
| Participants | n = 40 (15 men, 25 women) healthy, normosmic individuals, age: 19–35 years | n = 20 (9 men, 11 women) healthy, normosmic individuals, age: 19–30 years |
|
Stimulation Flow‐olfactometer OM2s |
Stimuli: CO2 35% v/v, H2S 5 ppm, PEA 50% v/v Stimulus duration: 400 ms ISI: 600, 1200, 1600, 3000, 5000, 6000, 7000, 9000 ms ITI: at least 30 s Repetitions: 40 times for each stimulus quality, a total of 120 stimulus pairs Estimation of intensity and hedonic valence after each stimulus pair |
Stimuli: H2S 5 ppm Stimulus duration: 400 ms ISI: 600, 1200 or 1600 ms ITI: at least 30 s Repetitions: 80 times Estimation of intensity and hedonic valence after each stimulus pair |
| EEG Recordings | Elektrode‐Position: Fz, Cz, Pz, C3 und C4, M2, M1, Fp2 | |
| Test Duration | 2 h and 20 min | 1 h and 30 min |
2.2. Stimulation
A computer‐controlled OM2s flow olfactometer (Burghart, Holm, Germany) was used to present the repeated chemosensory stimuli in both parts of the experiment. Constant odorless air was applied at a comfortable temperature of 37°C and delivered intranasally at 80% relative humidity with a flow rate of 8.8 L/min. In the first part, carbon dioxide (CO2, 35% v/v) as trigeminal, H2S (5 ppm) as unpleasant, and PEA (50% v/v) as pleasant stimulus were used [11]. The stimuli were presented with eight randomly selected different ISIs (600, 1200, 1600, 3000, 5000, 6000, 7000, 9000 ms). The stimulus sequence changed when a new stimulus quality was applied. Each stimulus was presented five times in combination with each ISI, with a stimulus duration of 400 ms, resulting in 40 repetitions for each stimulus quality and a total of 120 stimulus pairs. After 20 stimulus pairs of one stimulus quality, the side of intranasal stimulation was switched randomly. Participants were asked to press a button immediately as soon as they perceived a new stimulus.
Intranasal first stimulus presentation was inspiration‐synchronized using the LabVIEW application “ErfassungAtemzug_duotrig.vi” (developed by H. Lehnich: source code attached in Figures S1 and S2), which allows standardized recording of respiration and stimulus triggering (Figure 1) [12]. For this purpose, subjects breathed orally through a silicone tube, maintaining controlled sufficient velopharyngeal closure. A pressure sensor in the tube recorded pressure changes during inspiration and expiration. The intertrial intervals (ITIs) were set to 30 s waiting time in the LabVIEW program but varied depending on the regularity of the subjects' following breathing (coefficients of variation for inspiration and expiration duration of five continuous breaths: ≤ 0.2). If both experimental conditions were met, the flow olfactometer released a stimulus intranasally through a short nasal cannula beyond the nasal vestibule without the subject perceiving any pressure fluctuations during the 20 ms switching process. Sounds associated with switching of the stimulus were masked by white noise with 50 dB.
FIGURE 1.
Example of a breathing pattern during stimulation by H2S with an interstimulus interval of 1.2 s. Illustration of the triggering of twin stimulation and perception. Breathing is regular before stimulation and becomes highly irregular afterward.
2.3. Electroencephalography
In part 2 brain activity was recorded using EEG with the BrainVision Recorder (Brain Products GmbH, Gilching, Germany) and self‐adhesive EEG electrodes (Neuroline 270, Ambu GmbH, Bad Nauheim, Germany). Five electrodes were placed according to the international 10–20 system at positions Fz, Cz, Pz, C3, and C4, with two additional electrodes placed on the right (M2) and left (M1) mastoid process. An eighth electrode was put at position Fp2 to monitor vertical eye movements. EEG segments were recorded for 3500 ms with an additional pre‐trigger period of 500 ms. The sampling rate was set to 500 Hz.
After a familiarization period, during which the participants were informed about the procedure, gave their written informed consent and completed a medical history questionnaire, the EEG electrodes were attached, and the oral tube and nasal cannula were fitted. The experiment began with a trial run to ensure that the participants understood the procedure, the nasal cannula was correctly positioned, and the EEG was being recorded accurately.
Measurements were taken exclusively with H2S while recording EEG. A total of 80 H2S stimuli were presented intranasally as twin stimuli with the selected ISI. After 40 stimuli, the other side of the nose was stimulated in a randomized manner.
2.4. Data Analysis
EEG data were analyzed using Brain Vision Analyzer software version 2.1 (Brain Products GmbH, Gilching, Germany). Data were first filtered and segmented, then a baseline correction (−500 to 0 ms) was applied. The program automatically rejected blink artifacts measured using Fp2 above an amplitude of 50 μV. EEG artifacts, inspiratory stimulation and velopharyngeal closure were online visually verified. The CSERPs were averaged and evaluated using a one‐way ANOVA according to the perceptions of the twin stimuli (CSERP condition: 0 = perceived first stimulus, 1 = non‐perceived second stimulus, and 2 = perceived second stimulus). The presence of a CSERP was determined by visual inspection. If a CSERP was detected, the latencies N1 (largest negative peak between 200 and 700 ms) and P2 (largest positive peak between 300 and 800 ms), as well as the N1P2 amplitude difference, were recorded for the first stimulus. CSERPs in response to the second stimulus, considering the corresponding ISI and after adjustment based on known values for latencies N1 and P2 and the N1P2 amplitude [11], were recorded in the respective time windows.
2.5. Statistical Analysis
The statistical analysis of the collected data was performed using IBM SPSS Statistics 27 (IBM Deutschland GmbH, Böblingen, Germany). In the first part of the experiment, mixed linear models were used to account for correlations within individual participants. The dependent variable was the individual response rate. Independent factors were the duration of the ISI and the quality of the chemosensory stimulus. The influences of the ISI duration were analyzed at the eight different levels. Helmert contrasts were used to examine the monotonic increase in the proportion of correct responses until a plateau was reached. Seven multiple tests were used to compare the mean of one factor level with the mean of all subsequent factor levels (larger ISIs). To maintain an overall significance level of p = 0.05, individual comparisons were performed at a significance level of p = 0.05/7 after Bonferroni correction. Results for latencies N1 and P2, and amplitude N1P2 were examined by one‐way ANOVA between three CSERP‐conditions. In post hoc test the significance was adjusted by Tamhane. The effect size was rated according to Cohen 1998.
3. Results
3.1. Part 1 of the Experiment—Subjective Discrimination of Olfactory and Trigeminal Twin Stimuli
Pairs of CO2 stimuli were perceived as separate more often (85.4%) than H2S (75.2%) and PEA (62.1%) across all ISIs. H2S stimuli were discriminated with a higher success rate than PEA stimuli (Table 2, Figure 2). The highest plateau in terms of response rate was observed with CO2 stimulation, while the lowest plateau was observed with PEA stimulation. Pairwise Bonferroni‐corrected comparisons were performed to illustrate increases between ISIs and to identify differences between stimulus qualities. Analysis of each ISI revealed significant differences for very short ISIs. At ISIs of 600, 1200, and 1600, CO2 stimulus pairs were perceived as separate significantly more often than pairs of H2S and PEA stimuli. Even at ISIs of 3000 to 9000, CO2 twin stimuli were perceived separately significantly more often than PEA twin stimuli. At ISIs of 1600, 3000, 5000, 7000 and 9000 ms H2S twin stimuli were also perceived separately significantly more often than PEA twin stimuli. Using Helmert contrasts, the response rates of two consecutive ISIs were determined for each individual stimulus type. For CO2, a significant increase in the response rate was observed when increasing the ISI from 600 to 1200 ms (p < 0.001) and from 1200 to 1600 ms (p = 0.001). For H2S and PEA, a significant increase in response rate was observed when the ISI was increased from 600 to 1200 ms (p < 0.001), from 1200 to 1600 ms (p < 0.001), and from 1600 to 3000 ms (p < 0.001). All other response rates for the individual stimulants did not differ significantly (Table 2).
TABLE 2.
Part 1 of the experiment: Mean response rates (%) for all ISIs and the three stimulants CO2, H2S, and PEA: Phenylethyl Alcohol.
| ISI [ms] | CO2 | H2S | PEA |
|---|---|---|---|
| Response rate [%] | Response rate [%] | Response rate [%] | |
| 600 | 60.0 | 30.0 | 21.0 |
| 1200 | 81.0 | 53.5 | 42.0 |
| 1600 | 90.0 | 71.5 | 56.0 |
| 3000 | 90.0 | 90.5 | 71.5 |
| 5000 | 91.5 | 91.0 | 74.5 |
| 6000 | 90.5 | 88.0 | 78.5 |
| 7000 | 92.5 | 90.5 | 78.0 |
| 9000 | 88.0 | 87.0 | 75.0 |
| Total | 85.4 | 75.2 | 62.1 |
FIGURE 2.
Comparison of the response rates depending on the stimulants CO2, H2S, and phenylethyl alcohol (PEA); error bar chart for the response rates [%] with mean and 95% CI as a function of ISIs [ms]. ■ = CO2, ▼ = H2S, and ▲ = PEA.
3.2. Part 2 of the Experiment—Psychophysical and Electrophysiological Discrimination of H2 S Twin Stimuli
Of the 1600 H2S twin stimuli a total of 1527 (95%) were perceived and rated. Of these, 1182 (74%) were perceived separately, while 345 (22%) were perceived as a single stimulus. In 73 cases (4.6%), neither the first nor the second of the twin stimuli was perceived.
The first stimulus evoked CSERPs in all subjects, at least after stimulation of one nostril. The individual detection rate was 100%. When subjects perceived the second of the twin stimuli, a CSERP could be detected in all subjects at least at one recording position (Table 3). The individual detection rate was also 100%. When subjects did not perceive the second of the twin stimuli, the CSERP detection rate was 70% (Table 3). The one‐way ANOVA showed no significant differences in CSERP latencies between the three CSERP conditions (Figures 3 and 4). In contrast, a significantly higher amplitude was observed for CSERP Condition 1; however, with a small effect size (Table 4).
TABLE 3.
H2S‐CSERP by perceived and non‐perceived second stimulus of the twin stimuli (right or left nostril) with recording positions and the numbers of artifact‐free averaged EEG segments in brackets.
| Participant‐ID (ISI [ms]) | oCSERP by perceived second stimulus of the twin stimuli, stimulation: right or left; Detected at recording position (Number of averaged EEG segments) | oCSERP by non‐perceived second stimulus of the twin stimuli, stimulation: right or left; Detected at recording position (Number of averaged EEG segments) |
|---|---|---|
| 3 (600) |
Right: Fz, Cz, C3, C4, Pz (25) Left: Fz, Cz, C3 (14) |
Right: Fz, Cz, C3, C4, Pz (3) Left: Fz, Cz, C3 (9) |
| 11 (600) |
Right: Fz, Cz (30) Left: no ocserp |
Right: no oCSERP Left: no oCSERP |
| 24 (600) |
Right: Fz, Cz, C3, C4, Pz (24) Left: no oCSERP |
Right: no oCSERP Right: no oCSERP |
| 32 (600) |
Right: Fz (17) Left: Fz, Cz, C3, C4, Pz (5) |
Right: Fz, Cz, C3, C4, Pz (8) Left: Fz, C3 (11) |
| 38 (600) |
Right: Fz, Cz, C3, C4, Pz (22) Left: no oCSERP |
Right: Fz, C3 (3) Left: Fz, Cz, C3, C4, Pz (12) |
| 40 (600) |
Right: Fz, Cz, C3 (15) Left: Fz, Cz, C3, C4, Pz (17) |
Right: Fz, Cz, Pz (10) Left: no oCSERP |
| 5 (1200) |
Right: Fz, Cz, C3, C4, Pz (29) Left: Fz, Pz (22) |
Right: Fz (6) Left: no oCSERP |
| 9 (1200) |
Right: no oCSERP Left: Pz (25) |
Right: no oCSERP Left: Fz, Cz, C3 (4) |
| 10 (1200) |
Right: Fz, Cz, C3, C4, Pz (24) Left: Fz, Cz, C3, C4, Pz (24) |
Right: no oCSERP Left: Fz, Cz, C3, C4, Pz (8) |
| 14 (1200) |
Right: Fz, Cz, C3, C4, Pz (10) Left: Fz, Cz, C3, C4, Pz (18) |
Right: no oCSERP Left: Fz, Cz, C3, C4, Pz (9) |
| 16 (1200) |
Right: Fz, Cz, C3, C4, Pz (24) Left: Fz, Cz, C3, C4, Pz (21) |
Right: no oCSERP Left: no oCSERP |
| 21 (1200) |
Right: no oCSEP Left: Fz, Cz, C3, C4, Pz (31) |
Right: no oCSERP Left: no oCSERP |
| 25 (1200) |
Right: Fz, Cz, C3, C4, Pz (25) Left: Cz, C3, C4, Pz (23) |
Right: Fz, Cz, C3, C4, Pz (7) Left: Cz, C3, Pz (2) |
| 26 (1200) |
Right: no oCSERP Left: Fz, C3, C4 (27) |
Right: no oCSERP LEFT: no oCSERP |
| 29 (1200) |
Right: Cz, C3, Pz (22) Left: Fz, Cz, C3, C4, Pz (32) |
Right: no oCSERP Left: no oCSERP |
| 33 (1200) |
Right: Fz, Cz, C3, C4, Pz (28) Left: Fz, C3, C4 (27) |
Right: Fz, Cz, C3, C4, Pz (28) Left: Cz (4) |
| 34 (1200) |
Right: Cz, C3, C4, Pz (18) Left: C3 (17) |
Right: no oCSERP Left: Fz, Cz, C3, Pz (9) |
| 37 (1200) |
Right: Fz, Cz, C3, C4, Pz (7) Left: Fz, Cz, C3, C4, Pz (22) |
Right: Fz, C3, Pz (17) Left: Fz, Cz, C4, Pz (8) |
| 39 (1200) |
Right: Fz (21) Left: Fz, Cz, C4 (14) |
Right: Fz (11) Left: Fz, Cz, C4 (12) |
| 15 (1600) |
Right: Fz, Cz, C3, C4, Pz (27) Left: Fz, Cz, C3, C4, Pz (25) |
Right: Fz, Cz, C3, C4, Pz (7) Left: no oCSERP |
FIGURE 3.
Mean N1 latencies of olfactory chemosensory event‐related potentials (oCSERPs) for the first and second stimuli of the paired‐stimulus paradigm across different interstimulus intervals (ISIs), pooled across both nostrils. Points indicate means; error bars represent ±1 standard deviation.
FIGURE 4.
Mean P2 latencies of olfactory chemosensory event‐related potentials (oCSERPs) for the first and second stimuli of the paired‐stimulus paradigm across different interstimulus intervals (ISIs), pooled across both nostrils. Points indicate means; error bars represent ±1 standard deviation.
TABLE 4.
Result of the one‐way ANOVA (N1, P2, N1P2) between the CSERP‐conditions 0 (perceived 1st of the twin stimuli), 1 (non‐perceived 2nd of the twin stimuli), 2 (perceived 2nd of the twin stimuli) adjusted by Bonferroni with significance and effect size Eta2.
| Sum of squares | df | Mean of the squares | F | Sign. | Eta2 | ||
|---|---|---|---|---|---|---|---|
| Mean latency N1 | Between the groups | 139599.734 | 2 | 69799.867 | 2.787 | 0.063 | 0.01 |
| Within the groups | 8914375.013 | 356 | 25040.379 | ||||
| Total | 9053974.747 | 358 | |||||
| Mean latency P2 | Between the groups | 201518.828 | 2 | 100759.414 | 2.561 | 0.079 | 0.02 |
| Within the groups | 14047435.503 | 357 | 39348.559 | ||||
| Total | 14248954.331 | 359 | |||||
| Mean amplitude N1P2 | Between the groups | 923.658 | 2 | 461.829 | 10.047 | < 0.001 | 0.05 |
| Within the groups | 16594.616 | 361 | 45.968 | ||||
| Total | 17518.274 | 363 | |||||
4. Discussion
Chemosensory stimulus quality was found to influence stimulus perception. CO2 twin stimuli were discriminated significantly more often than twin stimuli of either odorant. H2S, selected as an unpleasant olfactory stimulus, was discriminated significantly more often than the pleasant stimulus PEA. From an evolutionary perspective, the sense of smell serves a critical role in detecting and evaluating potential hazards [13], which may explain the higher discrimination rates observed for unpleasant olfactory and trigeminal chemosensory intranasal stimuli.
In the first part of the experiment, CO2 reached its maximum trigeminal perception plateau (90%) for separate stimuli already at an ISI of 1600 ms. For both odorants their maximum was only reached at an ISI of 3000 ms. The maximum H2S‐perception plateau for separate stimuli was 90%, too, and lower at 72% for PEA. In the reference study, the three stimulants had lower perception plateaus, which were only reached with longer ISIs. The perception rates of the three stimuli types were also smaller in previous studies with stimulation independent of the respiratory cycle. It may therefore be conceivable that sufficiently high perception rates and reduction of examination time with inspiration‐synchronous stimulation could be achieved in clinical practice.
In the second part of the experiment 74% of the H2S twin stimuli were perceived separately. In the reference study, the percentage was considerably lower (56%), presumably due to respiratory asynchronous stimulation (see figure 2 in [1]) and although the ISI used was larger. In our study, with its very short ISIs, the second of the twin stimuli also mostly fell within inspiration. Most studies on the recording of olfactory CSERP used ISIs > 30 s [11]. Long examination times are usually associated with a decrease in vigilance, which in turn can result in increased background EEG activity that may affect CSERP components [14]. Examination time depends mainly on the length of the ISI, so it would be desirable to shorten it to obtain results faster. It has already been shown that reducing the ISI from 30 s to an average of 15 s can save 40% of the examination time [15]. Very short ISIs used in the present study additionally reduced the examination time. The long ITI fluctuated by > 30 s, depending on the time required to restore regular resting breathing. Subsequent studies should also investigate the effect of shorter ITIs by omitting the 30‐s waiting period as a trigger criterion. Only regular breathing over 5 breaths had to be established. Computer‐assisted stimulus delivery was verified to occur during inspiration or immediately prior to it, at the end‐expiratory resting position, in 76.5% of our archived respiratory files (N = 3263). In the remaining 23.5% of cases, stimuli were delivered during the preceding expiration. Reducing that rate would require participants to maintain an even more regular breathing pattern. Conversely, using a lower respiratory coefficient of variation as the trigger criterion would lengthen the experiment.
Our and Schriever et al. equal repetition rate led to an increase in examination time, too. However, the high number of repetitions improves the otherwise very low signal‐to‐noise ratio in CSERP recordings [16]. The oCSERP detection rate per subject was 100%. This is considered an indicator of intact olfactory function [17]. For normosmic individuals, other studies report the following detection rates: H2S‐CSERP 84% [18], 94% [19], and 93% [20]. In this context, a previous study reported that their CSERPs with an off‐line grouping of stimulation to the respiratory phase showed significantly larger N1 amplitudes during inspiration than during expiration for both CO2, H2S, and PEA [10]. Whether our CSERP detection rate is actually higher with inspiratory‐triggered first H2S stimuli than with standardized respiratory‐asynchronous triggering should be verified in a larger number of normosmics.
The H2S‐CSERPs of the second part of our experiment and the reference study are difficult to compare due to differing stimulus qualities and ISIs. The second of the twin stimuli mostly occurred in our study after very short ISIs during inspiration. The perceived separate stimuli did not differ in latency from those of the unperceived stimuli, as measured by the CSERP parameters. However, the amplitude value of the non‐perceived stimuli was larger. One explanation for the larger amplitudes observed in the case of non‐perceived stimuli could be that the P2 peak of CSERP reflects both the novelty of a stimulus and the cognitive load. Cerebral activation of secondary and tertiary olfactory areas is high in response to unpleasant stimuli, especially when the decision for or against stimulus perception is retrieved 80 times immediately after each separate stimulus, as in our study. Whitcroft et al. [21] described a similar finding in hyposmic patients. They observed an increased N1P2 amplitude in their PEA‐CSERPs compared to normosmic individuals likely because the hyposmic participants had a very high motivation to detect olfactory stimuli, similar to our test participants in order to detect the second of the twin stimuli after very short latencies.
The most notable difference in the experimental design of the present study compared with previous studies was the use of inspiration‐synchronous stimulation. The role of respiration in olfactory processing has long been recognized. Lorig et al. hypothesized that the respiratory cycle might influence central nervous system olfactory processing [22]. Various studies have investigated olfactory function under different conditions. In some cases, nasal stimulation was performed during voluntary, albeit uncontrolled, velopharyngeal closure or during normal mouth breathing without velopharyngeal closure [4, 5, 6, 7, 8, 9, 11, 12, 13, 14, 15, 22, 23, 24, 25, 26, 27, 28]. A 2011 study by Noguchi et al. [26] investigated how quickly and accurately individuals can distinguish different odors within a single breath. Odorants were presented at various time intervals following the onset of inspiration. The lowest perception threshold was observed when stimuli were delivered 0.2 s after the inspiration onset. These findings suggest that the brain can detect chemical changes in inhaled air within milliseconds, challenging the traditional view of slow odor processing [29, 30]. Thesen and Murphy found that P3 latencies were shorter when breathing under natural conditions than when breathing with velopharyngeal closure. They discussed that “in everyday smelling… odour perception rarely occurs independently of inhalation” [27]. It can be assumed that natural, active, inspiration‐synchronous olfactory stimulation leads to a priming of the olfactory system [10]. This would facilitate a phylogenetically advantageous, rapid stimulus evaluation. Passive, inspiration‐synchronous chemosensory stimulation, which is triggered by oral inspiration during controlled sufficient velopharyngeal closure, is a new experimental approach to studying natural, orthonasal, inspiration‐related chemosensory stimulation in humans. This approach highlights the rapid processing of olfactory stimuli with equal concentrations, durations, and qualities during inspiration.
5. Conclusion
When chemosensory stimulation was predominantly synchronized with inhalation, paired stimuli—even those with very short interstimulus intervals—were more frequently perceived as separate events than when stimulation was asynchronous with inhalation.
All participants exhibited at least unilateral H2S‐evoked chemosensory event‐related potentials (CSERPs) in response to perceived both the first and second of the paired olfactory stimuli at very short ISIs.
Funding
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1: User interface for the LabVIEW program “ErfassungAtemzug_duotrig.vi.”
Figure S2: Source code of the LabVIEW program “ErfassungAtemzug_duotrig.vi.”
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Schriever V. A., Frenzel C., Wernecke S., et al., “Olfactory Speed – Temporal Odor Processing of Paired Stimuli,” Neuroscience 295 (2015): 72–79. [DOI] [PubMed] [Google Scholar]
- 2. Wang L., Walker V. E., Sardi H., et al., “The Correlation Between Physiological and Psychological Responses to Odour Stimulation in Human Subjects,” Clinical Neurophysiology 113, no. 4 (2002): 542–551. [DOI] [PubMed] [Google Scholar]
- 3. Croy I., Maboshe W., and Hummel T., “Habituation Effects of Pleasant and Unpleasant Odors,” International Journal of Psychophysiology 88, no. 1 (2013): 104–108. [DOI] [PubMed] [Google Scholar]
- 4. Croy I., Krone F., Walker S., and Hummel T., “Olfactory Processing: Detection of Rapid Changes,” Chemical Senses 40, no. 5 (2015): 351–355. [DOI] [PubMed] [Google Scholar]
- 5. Sobel N., Prabhakaran V., Desmond J. E., et al., “Sniffing and Smelling: Separate Subsystems in the Human Olfactory Cortex,” Nature 392, no. 6673 (1998): 282–286. [DOI] [PubMed] [Google Scholar]
- 6. Alahäivälä A.‐L. I., Thaploo D., Wein S., et al., “Inhalation‐Modulated Detection of Olfactory BOLD Responses in the Human Brain,” Frontiers in Neuroimaging 2 (2023): 1260893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Zelano C., Jiang H., Zhou G., et al., “Nasal Respiration Entrains Human Limbic Oscillations and Modulates Cognitive Function,” Journal of Neuroscience 36, no. 49 (2016): 12448–12467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Perl O., Ravia A., Rubinson M., et al., “Human Non‐Olfactory Cognition Phase‐Locked With Inhalation,” Nature Human Behaviour 3, no. 5 (2019): 501–512. [DOI] [PubMed] [Google Scholar]
- 9. Arshamian A., Iravani B., Majid A., and Lundström J. N., “Respiration Modulates Olfactory Memory Consolidation in Humans,” Journal of Neuroscience 38, no. 48 (2018): 10286–10294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Haehner A., Gruenewald G., DiBenedetto M., and Hummel T., “Responses to Olfactory and Intranasal Trigeminal Stimuli: Relation to the Respiratory Cycle,” Neuroscience 175 (2011): 178–183. [DOI] [PubMed] [Google Scholar]
- 11. Hummel T., Klimek L., Welge‐Lüssen A., et al., “Chemosensorisch Evozierte Potential zur Klinischen Diagnostik von Riechstörungen,” HNO 48 (2000): 481–485. [DOI] [PubMed] [Google Scholar]
- 12. Lehnich H. and Gudziol H., “Erfassung des Atemmusters bei Geruchs‐ und Geschmacksreizen,” In. Virtuelle Instrumente in der Praxis, Rahman J. and Heinze R., eds. (VDE Verlag GmbH, 2011), 356–361. [Google Scholar]
- 13. Stevenson R. J., “An Initial Evaluation of the Functions of Human Olfaction,” Chemical Senses 35, no. 1 (2010): 3–20. [DOI] [PubMed] [Google Scholar]
- 14. Hummel T. and Kobal G., “Chemosensory Event‐Related Potentials to Trigeminal Stimuli Change in Relation to the Interval Between Repetitive Stimulation of the Nasal Mucosa,” European Archives of Oto‐Rhino‐Laryngology 256, no. 1 (1999): 16–21. [DOI] [PubMed] [Google Scholar]
- 15. Schaub F. and Damm M., “A Time‐Saving Method for Recording Chemosensory Event‐Related Potentials,” European Archives of Oto‐Rhino‐Laryngology 269, no. 10 (2012): 2209–2217. [DOI] [PubMed] [Google Scholar]
- 16. Boesveldt S., Haehner A., Berendse H. W., and Hummel T., “Signal‐To‐Noise Ratio of Chemosensory Event‐Related Potentials,” Clinical Neurophysiology 118, no. 3 (2007): 690–695. [DOI] [PubMed] [Google Scholar]
- 17. Kobal G. and Hummel T., “Olfactory (Chemosensory) Event‐Related Potentials,” Toxicology and Industrial Health 10, no. 4–5 (1994): 587–596. [PubMed] [Google Scholar]
- 18. Lotsch J. and Hummel T., “The Clinical Significance of Electrophysiological Measures of Olfactory Function,” Behavioural Brain Research 170, no. 1 (2006): 78–83. [DOI] [PubMed] [Google Scholar]
- 19. Gudziol H., Fischer J., Bitter T., and Guntinas‐Lichius O., “Chemosensory Event‐Related Brain Potentials (CSERP) After Strictly Monorhinal Stimulation,” International Journal of Psychophysiology 93, no. 3 (2014): 305–310. [DOI] [PubMed] [Google Scholar]
- 20. Gudziol H., Baldow E., Guntinas‐Lichius O., and Bitter T., “Erkennungsraten von Chemosensorisch Evozierten Potentialen bei Gesunden und Anosmischen Erwachsenen,” Laryngo‐Rhino‐Otologie 99, no. 11 (2020): 803–810. [DOI] [PubMed] [Google Scholar]
- 21. Whitcroft K., Whitcroft K. L., Aziz M., Croy I., Schriever V., and Hummel T., “Short Inter‐Stimulus Intervals Can Be Used for Olfactory Electroencephalography in Patients of Varying Olfactory Function,” Neuroscience 363 (2017): 26–33. [DOI] [PubMed] [Google Scholar]
- 22. Lorig T. S., Sapp A. C., Campbell J., and Cain W. S., “Event‐Related Potentials to Odor Stimuli,” Bulletin of the Psychonomic Society 31, no. 2 (1993): 131–134. [Google Scholar]
- 23. Prah J. D. and Benignus V. A., “Olfactory Evoked Responses to Odorous Stimuli of Different Intensities,” Chemical Senses 17, no. 4 (1992): 417–425. [Google Scholar]
- 24. Pause B. M., Sojka B., Krauel K., and Ferstl R., “The Nature of the Late Positive Complex Within the Olfactory Event‐Related Potential (OERP),” Psychophysiology 33, no. 4 (1996): 376–384. [DOI] [PubMed] [Google Scholar]
- 25. Murphy C., Nordin S., de Wijk R. A., Cain W. S., and Polich J., “Olfactory‐Evoked Potentials: Assessment of Young and Elderly, and Comparison to Psychophysical Threshold,” Chemical Senses 19, no. 1 (1994): 47–56. [DOI] [PubMed] [Google Scholar]
- 26. Noguchi D., Sugimoto S., Bannai Y.et al., “Time Characteristics of Olfaction in a Single Breath,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (Association for Computing Machinery, 2011), 83–92. [Google Scholar]
- 27. Thesen T. and Murphy C., “Age‐Related Changes in Olfactory Processing Detected With Olfactory Event‐Related Brain Potentials Using Velopharyngeal Closure and Natural Breathing,” International Journal of Psychophysiology 40, no. 2 (2001): 119–127. [DOI] [PubMed] [Google Scholar]
- 28. Nguyen D., Nguyen D. T., Rumeau C., Gallet P., and Jankowski R., “Olfactory Exploration: State of the Art,” European Annals of Otorhinolaryngology, Head and Neck Diseases 133, no. 2 (2016): 113–118. [DOI] [PubMed] [Google Scholar]
- 29. Sela L. and Sobel N., “Human Olfaction: A Constant State of Change‐Blindness,” Experimental Brain Research 205, no. 1 (2010): 13–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Spence C., McGlone F., Kettenmann B., and Kobal G., “Attention to Olfaction: A Psychophysical Investigation,” Experimental Brain Research 138 (2001): 432–437. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: User interface for the LabVIEW program “ErfassungAtemzug_duotrig.vi.”
Figure S2: Source code of the LabVIEW program “ErfassungAtemzug_duotrig.vi.”
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.