Background sound modulates the performance of odor ... - Springer Link

Exp Brain Res (2011) 212:305–314 DOI 10.1007/s00221-011-2729-5

R ES EA R C H A R TI CLE

Background sound modulates the performance of odor discrimination task Han-Seok Seo · Volker Gudziol · Antje Hähner · Thomas Hummel

Received: 5 December 2010 / Accepted: 9 May 2011 / Published online: 21 May 2011 © Springer-Verlag 2011

Abstract Even though we often perceive odors in the presence of various background sounds, surprisingly little is known about the eVects of background sound on odor perception. This study aimed to investigate the question whether background sound can modulate performance in an odor discrimination task. In Experiment 1, participants were asked to perform the odor discrimination task while listening to either background noise (e.g., verbal or nonverbal noise) or no additional sound (i.e., silent condition). Participants’ performance in the odor discrimination task was signiWcantly deteriorated in the presence of background noise compared with in the silent condition. Rather, the detrimental eVect of verbal noise on the task performance was signiWcantly higher than that of non-verbal noise. In Experiment 2, participants were asked to conduct the odor discrimination task while listening to either background music (Mozart’s sonata for two pianos in D major, K448) or no additional sound (silent condition). Background music relative to silent condition did not signiWcantly alter the task performance. In conclusion, our Wndings provide new empirical evidence that background sound modulates the performance in an odor discrimination task. Keywords Auditory–olfactory integration · Background noise · Background music · Odor discrimination task · Task performance

H.-S. Seo · V. Gudziol · A. Hähner · T. Hummel (&) Smell and Taste Clinic, Department of Otorhinolaryngology, University of Dresden Medical School, Fetscherstrasse 74, 01307 Dresden, Germany e-mail: [email protected]

Introduction We often smell odors while being exposed to various background sounds. For instance, in a metropolitan street, people experience exhaust fumes from automobiles while hearing traYc sounds (e.g., the sound of horns or car engines). Another example is an experience of crowd party where people smell many diVerent odors (e.g., food odors, body odors, and ambient fragrances) while hearing background music or listening to a conversation. Despite its high occurrence, surprisingly little is known about an auditory–olfactory integration. Previous studies have reported that auditory cues can inXuence on taste perception (Crisinel and Spence 2009; Woods et al. 2011) and food quality such as crispness and freshness (Vickers and Bourne 1976; Zampini and Spence 2004). For example, Woods et al. (2011) demonstrated that participants rated saltiness and sweetness as signiWcantly less intense when they ate food samples (e.g., biscuits, cracker, and cheese) in the presence of loud compared with quiet background noise. Only a few studies have addressed an association between olfactory and auditory stimuli in humans (Belkin et al. 1997; Spangenberg et al. 2005; Seo and Hummel 2011). SpeciWcally, Belkin et al. (1997) demonstrated that participants can consistently match certain auditory pitches with speciWc odorants based on odor quality. Focusing on contextual perception, Spangenberg et al. (2005) showed that congruency between ambient scent and background music in a retail setting can modulate consumers’ evaluation on the retail store, its merchandise, and the store environment. SpeciWcally, consumers evaluated the store more favorable when they were presented with a congruent combination of ambient Christmas scent and Christmas music than when presented with an incongruent combination (i.e., ambient Christmas scent and

123

306

non-Christmas music). Recently, Seo and Hummel (2011) presented empirical evidence that in comparison with incongruent sounds, congruent sounds can amplify pleasantness ratings of subsequently administrated odors. In addition, they showed that hedonic valence of preceding sounds can be transferred to pleasantness ratings of subsequently presented odors. For example, participants rated the presented odors as signiWcantly more pleasant following the presentation of pleasant sounds (e.g., baby laughing and jazz drum) than after unpleasant sounds (e.g., baby crying and screaming). Among many possible cases of olfactory–auditory integration, the current study attempted to highlight an inXuence of background sound on olfactory performance, in particular odor discrimination task. In fact, it is well known that background sound modulates task performance. Numerous studies have demonstrated that background noise interferes with the performance of cognitive tasks such as memory, recall, or reading comprehension (LeCompte et al. 1997; Furnham and Strbac 2002; Hygge 2003; Boman et al. 2005; Trimmel and Poelzl 2006; Cassidy and Macdonald 2007; Van Gerven et al. 2007; Wijayanto et al. 2009). DiVerent tasks mainly associated with the visual and/or the motor system have been employed in these studies to investigate the eVect of background sound on task performance (Crawford and Strapp 1994; Furnham and Bradley 1997; LeCompte et al. 1997; Boman et al. 2005; Trimmel and Poelzl 2006; Cassidy and Macdonald 2007; Van Gerven et al. 2007; Wijayanto et al. 2009). Yet little is known about the inXuences of background sound on the performance of tasks mainly associated with the sense of smell. Therefore, in Experiment 1, we sought to determine whether background sound can modulate olfactory task performance. Based on two assumptions: (1) background noise interferes with the performance of cognitive task (Hodge and Thompson 1990; LeCompte et al. 1997; Furnham and Strbac 2002; Hygge 2003; Boman et al. 2005; Trimmel and Poelzl 2006; Cassidy and Macdonald 2007; Van Gerven et al. 2007; Wijayanto et al. 2009) and (2) the cognitive function or education level is associated with olfactory performance (Larsson et al. 2004; Boesveldt et al. 2008; Seo et al. 2009; Hedner et al. 2010), we hypothesized that background noise disrupts the performance in an odor discrimination task. In particular, compared with other types of olfactory tasks, e.g., odor identiWcation or odor threshold, the odor discrimination task appears to demand cognitive function more strongly (Boesveldt et al. 2008; Hedner et al. 2010). A number of people work while listening to music. Indeed, many studies have reported the inXuence of background music on the performance in various tasks (Smith 1961; Crawford and Strapp 1994; Oldham et al. 1995; Turner et al. 1996; Furnham and Bradley 1997; Furnham and Allass 1999; Ho et al. 2007; Nantais and Schellenberg

123

Exp Brain Res (2011) 212:305–314

1999; Thompson et al. 2001; Furnham and Strbac 2002; Cassidy and Macdonald 2007; Bock 2010; Jäncke and Sandmann 2010). Among them, the “Mozart eVect,” Wrst reported by Rauscher and her colleagues, is frequently cited. SpeciWcally, Rauscher et al. (1993) demonstrated that participants’ performance on the standard IQ spatial reasoning task was improved immediately after listening to music composed by Mozart (especially Mozart’s sonata for two pianos in D major, K448), compared with their performance following the other two listening conditions: relaxation tape and silence. However, this “Mozart eVect” still remains controversial because the improved task performance is not consistently obtained (Chabris 1999; Steele et al. 1999; Pietschnig et al. 2010). In studies investigating the “Mozart eVect,” such a task (e.g., spatial–temporal task) was conducted after listening to the musical piece. However, the current study focused on the cross-modal integration (i.e., simultaneous performance of task in the presence of background music). Previous research has reported that in comparison with a silent condition, participants demonstrated signiWcantly worsened performance of tasks while listening to background music (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Strbac 2002; Cassidy and Macdonald 2007). In contrast, there are studies showing the opposite, namely that background music facilitates the task performance compared with a silent condition (Oldham et al. 1995; Turner et al. 1996; Ho et al. 2007). Rather, no signiWcant diVerence in the task performance between both conditions: background music and silence, has also been reported (Furnham and Allass 1999; Cassidy and Macdonald 2007; Jäncke and Sandmann 2010). This discrepancy in the results of earlier studies seems to be due to several factors such as type of task (Smith 1961; Oldham et al. 1995; Furnham and Bradley 1997; Furnham and Allass 1999; Cassidy and Macdonald 2007), type of music (Crawford and Strapp 1994; Turner et al. 1996; Cassidy and Macdonald 2007; Ho et al. 2007; Bock 2010), and the participant’s personality traits (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Allass 1999; Furnham and Strbac 2002; Cassidy and Macdonald 2007). In order to build on these Wndings, Experiment 2 was set out to examine the inXuence of background music on the task performance mainly associated with the sense of smell. Based on the notion that background music inXuences the performance of cognitive or occupational tasks (Crawford and Strapp 1994; Oldham et al. 1995; Furnham and Bradley 1997; Furnham and Allass 1999; Thompson et al. 2001; Furnham and Strbac 2002; Cassidy and Macdonald 2007; Ho et al. 2007; Bock 2010; Jäncke and Sandmann 2010), we wanted to test the hypothesis that background music (e.g., Mozart’s sonata) alters the performance of an odor discrimination task in comparison with a silent condition.

Exp Brain Res (2011) 212:305–314

307

Experiment 1

Odor discrimination task

Materials and methods

Odor discrimination task of the “SniYn’ Sticks” test (Hummel et al. 1997), well established as a clinical measure assessing olfactory function, was used. In the “SniYn’ Sticks” test, odorants were presented in felt-tip pens. The pen’s tampon is Wlled with liquid odorants or odorants dissolved in propylene glycol. The length and the inner diameter of the pens are approximately 14 and 1.3 cm, respectively. For odor presentation, the cap of the pen was removed by the experimenter and the tip of the pen was placed 2 cm in front of both nostrils. In the odor discrimination task, triplets of pens, with two containing the same and one containing a diVerent odorant, were presented in a randomized order. Using three alternative forced choice (3-AFC), participants were asked to select which of three pens smelled diVerently. The interval between presentations of individual pens was approximately 3 s. A total of 16 triplets were tested. Because one point was assigned for each correct answer, the participants’ scores ranged from 0 to 16. The interval between the presentations of odor triplets was 20–25 s.

Participants Thirty-eight healthy right-handed volunteers (29 women) with an age range between 19 and 40 years [mean age § standard deviation (SD) = 25.1 § 3.7 years] took part in Experiment 1. To minimize a possible inXuence of handedness on olfactory performance and brain response (BensaW et al. 2003; Royet et al. 2003), only right-handed volunteers took part in the experiment. In fact, BensaW et al. (2003) found that handedness inXuences the time taken to perform an odor familiarity task; for example, right-handers performed the familiarity judgment more quickly than left-handers. Using a functional magnetic resonance imaging, Royet et al. (2003) demonstrated a lateralized olfactory emotional processing in the ventral insula as a function of handedness. Handedness was determined using a translated version of the Edinburgh inventory (OldWeld 1971). Participants were recruited via leaXet. All participants conWrmed that they had no clinical history of major diseases and that they had normal senses of smell and hearing. All participants had no impairment in olfactory, auditory, or cognitive function on the basis of the following test results: the “SniYn’ Sticks” screening test (Burghart Instruments, Wedel, Germany; for details, see Hummel et al. 2001), the tuning fork test (Doyle et al. 1984), and the “Mini-Mental-State Examination” (Folstein et al. 1975), respectively. The experimental procedure was explained to all participants, and informed written consent was obtained. Participants were not informed about all details about the purpose of study prior to their participation. They were debriefed about these details after the completion of the study. Background sound and presentation To set out background noise (referred to “unwanted or irrelevant sound”) under this study, two auditory stimuli were used: the sound of crowed party (#258) obtained from a web provider of sound eVects (http://free-loops.com) and the sound of audio book (Mario Barth: Langenscheidt “Frau-Deutsch Deutsch Frau 2”, Sony Music Entertainment Germany GmbH, Germany). The sound of a crowded party (subsequently to be called “party sound”) presented various sounds produced in the crowed party. The sound of the audio book (subsequently to be called “audio book”) presented famous known German comedian’s humorous speech. Both sounds were provided via a headphone at a loudness of 70 dB. Finally, in the silent condition, no additional auditory stimulus was presented.

Procedure This study consisted of two sessions. To minimize a possibility that repeated trials of odor discrimination task inXuence its performance (i.e., learning eVect), the odor discrimination task was conducted over two sessions on diVerent days, with a maximum of 7 days in between. In each session, participants were asked to perform odor discrimination task two times: in the absence of background noise (subsequently to be called “silent”) and in the presence of background noise (subsequently to be called “noisy”). The order of silent and noisy conditions was randomly counterbalanced across participants. Moreover, within the noisy condition, the order of two background noises (i.e., “party sound” and “audio book”) were randomly counterbalanced across participants. For example, if the Wrst session consists of silent followed by party sound, the second session consists of audio book followed by silent condition. Participants were seated on a chair 55 cm in front of a computer monitor. Participants were instructed to wear headphones (MDR-A44L, Sony Corporation, Tokyo, Japan) and watch the computer monitor. In the original version of the odor discrimination task (Hummel et al. 1997), to prevent the visual detection of the target odor pen, participants are required to be blindfolded and experimenter says the number (e.g., “one,” “two,” or “three”) of pens while presenting each odorant pen. However, in this study, participants were required to watch the computer monitor where the number (e.g. “1,” “2,” or “3”) of the presented pen was shown. That

123

308

was because participants had trouble listening to the experimenter’s advice in the presence of background noise. Instead, to minimize the visual detection of the target pen with a cap colored in green, the cap (colored in green, red, or blue) of pens was concealed while the odorant pen was presented to the participants. The colored cap was easily concealed when the experimenter held the pen. The identical procedure was employed in the silent condition, even though participants could listen to the experimenter’s instructions. After sniYng all three odors one by one in the silent or noisy condition, participants were asked to indicate the number of the pen containing the diVerent odorant. In each session, the interval between silent and noisy conditions was 10–15 min. Data analyses Data analyses were performed by means of SPSS 17.0 for Windows™ (SPSS Inc, Chicago, IL, USA). To test the inXuences of two factors: (1) presence of background noise (i.e., “silent condition” vs. “noisy condition”) and (2) sessions conducted on two diVerent days (i.e., “party sound” session vs. “audio book” session) on the task performance, a two-way repeated measures analysis of variance (RMANOVA) was used. A possible interaction between two factors (i.e., “presence of background noise” by “sessions conducted on two diVerent days”) was also tested using the RM-ANOVA. In addition, using paired t tests, we compared the diVerences between mean scores of the silent and noisy conditions according to the type of background noise [i.e., (“silent”—“party sound”) vs. (“silent”—“audio book”)]. Furthermore, the frequency of participants who showed even, better, or worse performance in the noisy condition compared with the silent condition was examined using the McNemar test. Due to the low frequency of participants, the frequencies of participants who produced even or improved performance in the noisy condition (i.e., party sound or audio book) than in the silent condition were combined for statistical analyses. The alpha level was 0.05. Results The two-way RM-ANOVA revealed that the mean score of odor discrimination task was signiWcantly diVerent between the presence of background noise and its absence [F(1, 37) = 36.48, P < 0.001]. SpeciWcally, participants produced worsened performance in the odor discrimination task when presented with background noise (mean § SD = 11.9 § 1.8) compared with the situation when no additional auditory stimulus was presented (13.0 § 1.5). No signiWcant diVerence in the mean score of odor discrimination task exhibited between two sessions conducted on diVerent days: that is, party sound session vs. audio book session

123

Exp Brain Res (2011) 212:305–314

Fig. 1 Comparison of the mean scores of odor discrimination task in the “silent” and “noisy” conditions. Compared with the silent condition, the mean score of odor discrimination task was signiWcantly lower in the presence of background noise (e.g., “party sound” or “audio book”). In addition, comparing the diVerence of mean scores in the silent and noisy conditions in relation to the type of background noise, the score diVerence of the audio book session was signiWcantly higher than that of the party sound session. *SigniWcance at P < 0.05. The error bars represent the standard errors of the means

(P = 0.71). There was a signiWcant interaction between two factors: “presence of background noise” and “sessions” [F(1, 37) = 6.43, P = 0.02]. As shown in Fig. 1, the background noise-induced disturbance in the odor discrimination task appears to be more pronounced in the session using audio book than in the session using party sound as background noise. We compared the diVerence of mean scores between the silent and noisy conditions in relation to the type of background noise (i.e., non-verbal noise of “party sound” vs. verbal noise of “audio book”). The score diVerence of the audio book session (i.e., silent—audio book; mean § SD = 1.5 § 1.4) was about two times higher than that of the party sound session (i.e., silent—party sound; 0.7 § 1.6), as seen in Fig. 1. In other words, the detrimental eVect of verbal type noise (i.e., audio book) on the task performance was signiWcantly greater than that of nonverbal type of noise (i.e., party sound) [t(37) = ¡2.54, P = 0.02]. The McNemar test revealed that the frequency of participants who exerted poorer performance in the presence of background noise than in the absence of noise was signiWcantly diVerent in relation to the type of background noise: party sound and audio book (P = 0.02). SpeciWcally, compared with the silent condition, verbal noise of audio book yielded poorer performance of the participants more frequently (N = 30, 78.9%) than the non-verbal noise of party sound (N = 20, 55.6%). Discussion Our Wndings demonstrate, for the Wrst time, that background noise can conXict performance in a target task

Exp Brain Res (2011) 212:305–314

mainly associated with the sense of smell, as we hypothesized. This result is in line with the previous studies demonstrating that background noise interfered with results in cognitive or occupational tasks, which have been involved in the visual and/or motor system (Hodge and Thompson 1990; LeCompte et al. 1997; Furnham and Strbac 2002; Hygge 2003; Boman et al. 2005; Trimmel and Poelzl 2006; Cassidy and Macdonald 2007; Van Gerven et al. 2007; Wijayanto et al. 2009). Among the various olfactory tasks, an odor discrimination task seems to be highly dependent on cognitive abilities (Boesveldt et al. 2008; Hedner et al. 2010). For example, Hedner et al. (2010) reported that cognitive ability including executive function, episodic memory, and semantic memory is positively associated with proWciency in odor discrimination or odor identiWcation, but not with an odor threshold task. Accordingly, it can be thought that the background noise-induced worsened performance may be mediated by interrupted cognitive processes required to conduct the odor discrimination task. It is worth noting that the detrimental eVect of background noise on the task performance varied as a function of the background noise type. That is, participants showed more deleterious performance on the odor discrimination task while listening to verbal noise (i.e., audio book) than while listening to non-verbal noise (i.e., party sound). This result is in accordance with previous evidence demonstrating that the disrupting eVect of background noise varied depending on contextual variables of background noise (Crawford and Strapp 1994; LeCompte et al. 1997; Boman et al. 2005; Cassidy and Macdonald 2007; Van Gerven et al. 2007). For example, LeCompte et al. (1997) showed that meaningful speech interfered with performance of the recall task more strongly than meaningless speech or tone. Boman et al. (2005) reported similar Wndings that meaningful speech deteriorated participants’ performance in a reading comprehension test more robustly than road traYc noise. Accordingly, in the current study, it can be assumed that in comparison with party sound including little semantic cues, the comedian’s humorous speech gave more semantic information, which could draw participants’ attention from the target task more easily. Indeed, it was observed that not a few participants often laughed in the presence of the “audio book,” indicating that the participants attended to listening to the humorous speech rather than performing the odor discrimination task.

309

the task performance. It is assumed that background noise appears to exert a deleterious eVect on the task performance, whereas background music seems to produce not only worsened performance but also improved performance of the target task (see “Introduction”). Rather, while background oYce noise generally evokes a negative attitude due to its unpredictable and uncontrollable characteristics, background music may induce positive as well as negative mood (Furnham and Strbac 2002). Furthermore, no study, to our best knowledge, has been conducted to assess the inXuence of background music on the olfactory task performance. Therefore, in Experiment 2, we attempted to investigate whether background music (using the same piece as employed in the “Mozart eVect” by Rauscher et al. (1993)) can alter the performance in an odor discrimination task. Materials and methods Participants Thirty-six healthy right-handed volunteers (27 women) with an age range between 19 and 36 years (mean age § SD = 24.1 § 3.1 years) participated in Experiment 2. Among them, 14 volunteers (4 men) also participated in the Experiment 1. Experiment 2 was conducted one month after completion of Experiment 1. Handedness was determined using a translated version of the Edinburgh inventory (OldWeld 1971). All participants conWrmed that they had no clinical history of major diseases and that they had normal senses of smell and hearing. All participants had no impairments in olfactory, auditory, and cognitive functions based on the results of screening tests described in Experiment 1. The screening tests were not conducted for 14 participants who took part in Experiment 1. The experimental procedure was explained to all participants, and informed written consent was obtained. Participants were not informed about all details about the purpose of study prior to their participation. They were debriefed about these details after completion of the study. Background sound and presentation As background sound, Mozart’s musical piece (sonata for two pianos in D major, K448) (Mozart: Music for 2 pianos, Piano Duets, Philips) was employed. This background music was presented via a headphone at a loudness of 70 dB. In the silent condition, no additional auditory stimulus was presented.

Experiment 2 Odor discrimination task In contrast to the interrupting eVect of background noise on the performance of cognitive or occupational tasks, it still remains unclear whether background music inXuences on

The same task of odor discrimination in the “SniYn’ Sticks” test (Hummel et al. 1997) was conducted using the

123

310

identical procedure as in Experiment 1. BrieXy, triplets of pens with two containing the same and one containing a diVerent odorant were presented in a randomized order. Using 3-AFC, participants were required to select which of three pens smelled diVerently. The interval between the presentations of individual pens was approximately 3 s. A total of 16 triplets were tested. The interval between presentations of odor triplets was 20–25 s. Procedure Participants were asked to perform the odor discrimination task two times on the same day: in the absence of background music (subsequently to be called “silent”) and in the presence of background music (subsequently to be called “music”). The order of silent and music conditions was randomly counterbalanced across participants. The time interval between silent and music conditions was 10–15 min. The odor discrimination task was conducted under the identical procedure as described for Experiment 1. BrieXy, participants were seated on a chair 55 cm in front of a computer monitor and were instructed to wear headphones (MDR-A44L, Sony Corporation, Tokyo, Japan). As described in Experiment 1, instead of hearing the number of pens while presenting each odorant pen, participants were asked to watch the computer monitor where the number (e.g., “1,” “2,” or “3”) of the presented pen was shown. To minimize the visual detection of the target pen with a cap colored in green, the cap (colored in green, red, or blue) of pens was concealed while the odorant pen was presented to the participants. After sniYng all three odors one by one in the silent or music condition, participants were asked to indicate the number of the pen containing the diVerent odorant. In each session, the interval between silent and music conditions was 10–15 min. After experimental session, participants were asked to rate the background music pleasantness and arousal on two eleven-point scale ranging between ¡5 (extremely unpleasant/extremely calm) and +5 (extremely pleasant/extremely arousing), respectively. Data analyses Data analyses were conducted by means of SPSS 17.0 for Windows™ (SPSS Inc, Chicago, IL, USA). Because 14 volunteers took part in Experiment 1, a preliminary analysis using an independent t test was conducted to compare their performance (i.e., mean score) in the odor discrimination task with naïve participants’ task performance. To test whether participants’ better performance in the silent condition correlates with their better performance in the presence of background music, the Pearson’s correlation analysis was used. Using paired t tests, the mean scores of

123

Exp Brain Res (2011) 212:305–314

the task were compared between “silent” and “music” conditions to determine whether background music alters the performance in an odor discrimination task. In addition, Pearson and Kendall (tau_b) correlation analyses were used to test whether participants’ task performance correlates with the ratings of either pleasantness or arousal for the presented music. Due to the small size of data (e.g., between 11 and 13) in each case, the non-parametric correlation analysis (i.e., Kendall’s tau_b) was used in addition to the Pearson’s correlation analysis. The alpha level was 0.05. Results As mentioned above, because 14 volunteers had participated in Experiment 1, we compared their performance in the odor discrimination task with that of naïve participants (N = 22). A preliminary analysis using an independent t test revealed that the diVerence in mean scores in the silent and music conditions was not signiWcantly diVerent between two participants’ groups [t(34) = 1.04, P = 0.31]. Therefore, we considered two participants’ groups as one group in all subsequent analyses. Correlation analyses found a signiWcant relationship of individual scores in the odor discrimination task in the silent and music conditions (r(36) = 0.66, P < 0.001). That is, as participants showed better performance in the silent condition, they yielded better performance in the presence of background music. Among 36 participants, 36.1% (N = 13) displayed better performance in the odor discrimination task in the silent condition, whereas 33.3% (N = 12) showed better performance in the presence of background music. The others (30.6%, N = 11) demonstrated no diVerence in the performance of odor discrimination task in the silent and music conditions. Paired t tests revealed that the mean score of the odor discrimination task was not signiWcantly diVerent in the silent (mean § SD = 13.3 § 1.5) and music (13.3 § 1.8) conditions (P = 0.81). That is, on average, background music of Mozart’s piece showed neither enhanced nor reduced eVect on the odor discrimination task as compared with the silent condition. Participants judged the background music as being highly pleasant (3.6 § 1.8) and neither calm nor arousing (0.8 § 2.5). Figure 2 shows the relationships of individual performance of odor discrimination task in the presence of background music with either individual rating of pleasantness (a) or arousal (b) for the presented music. Pearson’s correlation analyses showed that individual score of odor discrimination task in the presence of background music correlated with neither individual rating of pleasantness [r(36) = 0.13, P = 0.45] nor arousal [r(36) = 0.04, P = 0.81]. However, as seen in Fig. 2a, an association between individual score of odor discrimination task in the music

Exp Brain Res (2011) 212:305–314

311

Fig. 2 The relationships of individual score of odor discrimination task in the presence of background music with either individual rating of pleasantness (a) or arousal (b) for the presented music. The “worse”, “even”, and “better” indicate participants who showed worse, even, and better performance in the presence of background music than in the silence, respectively. The numeral letters on the plot indicate the number of redundant data. For example, the “6” on a certain spot indicates that six data having the same value were redundant on the spot

condition and individual pleasantness rating of the music seems to be diVerent in relation to participants’ task performance. SpeciWcally, among the participants showing better performance in the music condition than in the silent condition, positive but not signiWcant correlation was found [tau_b(12) = 0.36, P = 0.18; c.f. Pearson’s correlation coeYcient r(12) = 0.74, P < 0.01]. In other words, among the participants who discriminated the odors more correctly in the presence of music, the more they liked the music, the higher their performance was in the presence of this music. In contrast, negative but not signiWcant correlation was observed among the participants showing worse performance in the presence of music than in its absence [tau_b(13) = ¡0.23, P = 0.35; c.f. Pearson’s correlation coeYcient r(13) = ¡0.21, P = 0.49]. Figure 2b presents an association between individual score of odor discrimination task in the music condition and individual arousal ratings of music, in relation to participants’ task performance. Among participants showing worse, even, or better performance in the music condition than in the silent condition, a negative [tau_b(13) = ¡0.07, P = 0.75; c.f. Pearson’s correlation coeYcient r(13) = ¡0.10, P = 0.76], positive [tau_b(11) = 0.18, P = 0.47; c.f. Pearson’s correlation coeYcient r(11) = 0.25, P = 0.47], or positive [tau_b(12) = 0.18, P = 0.50; c.f. Pearson’s correlation coeYcient r(12) = 0.16, P = 0.62] but not signiWcant relation was obtained, respectively. Discussion Contrary to our hypothesis, the results demonstrate no signiWcant inXuence of background music (especially Mozart’s sonata for two pianos in D major, K448) on the performance of odor discrimination task. However, it is interesting to note that the frequency of participants who showed better (33.3%), even (30.6%), or worse (36.1%)

performance in the presence of background music than in the silent condition was almost equally distributed. Although only one background music was used as auditory stimulus in the current study, this Wnding reXects that the modulating eVect of background music on the task performance may vary across participants. In fact, previous studies have demonstrated that background music relative to silence produces not only worsened performance (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Strbac 2002; Cassidy and Macdonald 2007) but also improved performance (Oldham et al. 1995; Turner et al. 1996; Ho et al. 2007); no signiWcant eVect of background music has been reported (Smith 1961; Furnham and Allass 1999; Cassidy and Macdonald 2007; Jäncke and Sandmann 2010). As mentioned in Introduction, this discrepancy of results across early studies may be due to diVerences in experimental design (e.g., type of task, type of background music, or participant’s personality traits). SpeciWcally, it has been reported that the modulating eVect of background music on the task performance varies depending on the type of task (Smith 1961; Oldham et al. 1995; Furnham and Bradley 1997; Furnham and Allass 1999; Cassidy and Macdonald 2007). For example, it is assumed that background music improves the performance of monotonous or routine tasks by reducing boredom and/or increasing positive mood, whereas the music impairs the performance of complex tasks by distracting attention from the target task (Smith 1961; Oldham et al. 1995; Furnham and Allass 1999; Cassidy and Macdonald 2007), which is, to some extent, in agreement with our results. SpeciWcally, when participants produced higher scores in the odor discrimination task in the silence, they also showed better performance in the presence of background music. In other words, the easier the task was for participants, the better the task performance became in the presence of background music.

123

312

One plausible explanation for the lack of signiWcance is the low task diYculty, which resulted in a ceiling eVect. That is, as seen in Fig. 2a, the odor discrimination task used in this study seems to be relatively easy for most participants, which subsequently may reduce the sensitivity to the background music.

General discussion The current Wndings add new evidence to a list reporting the inXuences of background sound on task performance. In earlier studies, researchers have employed cognitive or occupational tasks that have been mainly associated with the visual and/or the motor system. However, little is known whether background sound aVects olfactory task performance. Therefore, herein for the Wrst time, we report a modulating eVect of background sound on the performance of odor discrimination task. The main Wndings of the present study are as follows: 1. Compared with a silent condition, background noise interferes with the performance of the odor discrimination task. 2. The detrimental eVect by background noise on task performance varies as a function of the type of background noise. 3. Background music (speciWcally Mozart sonata for two pianos in D major, K448) relative to silence reveals no diVerence in the average performance in an odor discrimination task. Taken together, the results of this study support the notion that background noise generally interferes with task proWciency (LeCompte et al. 1997; Furnham and Strbac 2002; Hygge 2003; Boman et al. 2005; Trimmel and Poelzl 2006; Cassidy and Macdonald 2007; Van Gerven et al. 2007; Wijayanto et al. 2009). In addition, our results corroborate the previous Wndings that background music has little or no eVect on the task performance (Smith 1961; Furnham and Allass 1999; Cassidy and Macdonald 2007; Jäncke and Sandmann 2010). However, as addressed earlier, there is empirical evidence showing that background music not only can lessen the task performance (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Strbac 2002; Cassidy and Macdonald 2007) but also can improve the performance (Oldham et al. 1995; Turner et al. 1996; Ho et al. 2007) in various tasks. In addition, as commented in Experiment 2, the low task diYculty might diminish participants’ sensitivity to the background music. Therefore, careful consideration is needed to establish the inXuence of background music on the performance of odor discrimination task.

123

Exp Brain Res (2011) 212:305–314

Even though our Wndings present no or detrimental inXuence of background sound on the performance of odor discrimination task, it is worth noting that some participants showed better performance while listening to background sound relative to the silence (e.g., party sound and background music in Experiments 1 and 2, respectively). SpeciWcally, in Experiment 2, 33.3% (N = 12) showed better performance while listening to the background music than in the silent condition. One of plausible explanations for this improved eVect is that the inXuence of background sound on the task performance might be dependent on the participant’s personality traits (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Allass 1999; Furnham and Strbac 2002; Cassidy and Macdonald 2007; for review, Belojevic et al. 2003). Indeed, extroverts are more likely to work with the radio on than introverts (Furnham and Bradley 1997). In addition, many studies have demonstrated that compared with extroverts, introverts have higher sensitivity to noise and show less successful performance in various tasks while listening to a background sound (Crawford and Strapp 1994; Furnham and Bradley 1997; Furnham and Allass 1999; Furnham and Strbac 2002; Cassidy and Macdonald 2007). Rather, it is reported that as complexity of background music increases, extrovert’s performance on cognitive task is improved, whereas introvert’s performance is worsened (Furnham and Allass 1999; Cassidy and Macdonald 2007). It is tempting to speculate about a possible neural mechanism underlying the modulatory inXuence of background sound on the performance of odor discrimination task. There is a growing list of evidence for the integration of auditory and olfactory stimuli. SpeciWcally, recent studies found that auditory cortex responds to not only auditory stimulus, but also visual stimuli (Werner-Reiss et al. 2003; Brosch et al. 2005) or somatosensation (Brosch et al. 2005). Moreover, using the gerbil model, Budinger et al. (2006) demonstrated that the auditory cortex receives direct inputs not only from visual and multisensory thalamic and brainstem structures, but also from visual, somatosensory, and multisensory cortices. Further, the auditory cortex projects directly to primary olfactory, somatosensory, visual, and multisensory cortical and subcortical brain areas. More recently, using a mouse model, Wesson and Wilson (2010) revealed that 65 and 19% of olfactory tubercle single-units responded to odors and auditory tones, respectively. Moreover, 29% of single-units showed supra-additive or suppressive responses to the simultaneous presentation of odors and tones, suggesting that the olfactory tubercle may play an important role in modulating the auditory–olfactory integration. A functional magnetic resonance imaging study by Plailly et al. (2007) found that the brain areas underlying the feeling of familiarity to music or odors are overlapping in the left hemisphere, speciWcally the precuneus, the

Exp Brain Res (2011) 212:305–314

superior and inferior frontal gyri, the angular gyrus, the hippocampus, and the parahippocampal gyrus. In addition to the anatomical convergence of brain areas involving auditory or olfactory processing, the modulatory inXuence of background sound seems to be mediated in brain areas associated with cognitive processing in the odor perception. SpeciWcally, because the odor discrimination task used in this study is highly associated with participants’ cognitive function (Boesveldt et al. 2008; Hedner et al. 2010), it seems that brain areas (e.g., amygdala, hippocampus, and orbitofrontal cortex) responsible for odor attention, cognition, or memory (for review Gottfried 2006; Rolls and Grabenhorst 2008) may be associated with the modulatory eVect of background sound on the task performance. That is, it can be assumed that participants’ attention may be drawn away from the olfactory stimuli to the background sound, which subsequently may aVect participants’ performance in the odor discrimination task. Indeed, previous cross-modal studies (e.g., visual–auditory interactions) have demonstrated that modality-speciWc selective attention decreases not only task performance but also neural activity in the unattended sensory modality, for example, “cross-modal deactivations” (Hairston et al. 2008; Mozolic et al. 2008). The degree of cross-modal deactivations is associated with the diYculty of task given during the cross-modal condition (Hairston et al. 2008). For example, the cross-modal deactiviations signiWcantly increased as the task was more diYcult (Hairston et al. 2008). It may also be that the modulatory eVect of background sound on the performance in the odor discrimination task is mediated in brain areas (e.g., amygdala, insula, cingulate, and orbital-prefrontal cortices) associated with aVective processing (Dolan 2002). SpeciWcally, many studies have reported that background sound evokes positive or negative mood (Oldham et al. 1995; Furnham and Strbac 2002). Further, it is known that participants’ mood or emotional state also inXuences not only odor intensity, pleasantness, or response time (Chen and Dalton 2005; Seo and Hummel 2011), but also cognition, attention, and memory (Dolan 2002). Accordingly, a background sound-induced emotional status might modulate participants’ performance in the odor discrimination task via direct or indirect pathway. To summarize, the present Wndings provide, for the Wrst time, empirical evidence that background sound aVects the performance in an odor discrimination task. Further study is needed to investigate a possible inXuence of contextual variables (e.g., type of task, type of background music, and participant’s personality traits) on the modulatory eVect of background sound on participants’ performance in the odor discrimination task. Furthermore, a larger number of participants are necessary to generalize the impact of variables listed and the eVect of music liking on the task performance.

313 Acknowledgments This research was supported by a grant from the Centre National de la Recherche ScientiWque to T.H. (European associated laboratory; EAL 549, CNRS-TUD).

References Belkin K, Martin R, Kemp SE, Gilbert AN (1997) Auditory pitch as a perceptual analogue to odor quality. Psychol Sci 8:340–342 Belojevic G, Jakovljevic B, Slepcevic V (2003) Noise and mental performance: personality attributes and noise sensitivity. Noise Health 6:77–89 BensaW M, Rouby C, Farget V, Bertrand B, Vigouroux M, Holley A (2003) Perceptual, aVective, and cognitive judgments of odors: pleasantness and handedness eVects. Brain Cogn 51:270–275 Bock O (2010) Sensorimotor adaptation is inXuenced by background music. Exp Brain Res 203:737–741 Boesveldt S, Verbaan D, Knol DL, Visser M, van Rooden SM, van Hilten JJ, Berendse HW (2008) A comparative study of odor identiWcation and odor discrimination deWcits in Parkinson’s disease. Mov Disord 23:1984–1990 Boman E, Enmarker I, Hygge S (2005) Strength of noise eVects on memory as a function of noise source and age. Noise Health 7:11– 26 Brosch M, Selezneva E, Schelch H (2005) Nonauditory events of a behavioral procedure activate auditory cortex of highly trained monkeys. J Neurosci 25:6797–6806 Budinger E, Heil P, Hess A, Scheich H (2006) Multisensory processing via early cortical stages: connections of the primary auditory cortical Wled with other sensory systems. Neuroscience 143:1065– 1083 Cassidy G, MacDonald RAR (2007) The eVect of background music and background noise on the task performance of introverts and extraverts. Psychol Music 35:517–537 Chabris CF (1999) Prelude or requiem for the ‘Mozart eVect’? Nature 400:826–827 Chen D, Dalton P (2005) The eVect of emotion and personality on olfactory perception. Chem Senses 30:345–351 Crawford HJ, Strapp CM (1994) EVects of vocal and instrumental music on visuospatial and verbal performance as moderated by studying preference and personality. Person Individ DiV 16:237–245 Crisinel AS, Spence C (2009) Implicit association between basic tastes and pitch. Neurosci Lett 464:39–42 Dolan RJ (2002) Emotion, cognition, and behavior. Science 298:1191– 1194 Doyle PJ, Anderson DW, Pijl S (1984) The tuning fork: an essential instrument in otologic practice. J Otolaryngol 13:83–86 Folstein MF, Folstein SE, McHugh PR (1975) Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189–198 Furnham A, Allass K (1999) The inXuence of musical distraction of varying complexity on the cognitive performance of extroverts and introverts. Eur J Pers 13:27–38 Furnham A, Bradley A (1997) Music while you work: the diVerential distraction of background music on the cognitive test performance of introverts and extroverts. Appl Cogn Psychol 11:445– 455 Furnham A, Strbac L (2002) Music is as distracting as noise: the diVerential distraction of background music and noise on the cognitive test performance of introverts and extraverts. Ergonomics 45:203–217 Gottfried JA (2006) Smell: central nervous processing. Adv Otorhinolaryngol 63:44–69 Hairston WD, Hodges DA, Casanova R, Hayasaka S, Kraft R, Maldjian JA, Burdette JH (2008) Closing the mind’s eye: deactivation

123

314 of visual cortex related to auditory task diYculty. Neuroreport 19:151–154 Hedner M, Larsson M, Arnold N, Zucco GM, Hummel T (2010) Cognitive factors in odor detection, odor discrimination, and odor identiWcation tasks. J Clin Exp Neuropsychol 32:1062–1067 Ho C, Mason O, Spence C (2007) An investigation into the temporal dimension of the Mozart eVect: evidence from the attentional blink task. Acta Psychol 125:117–128 Hodge B, Thompson JF (1990) Noise pollution in the operating theatre. Lancet 335:891–894 Hummel T, Sekinger B, Wolf SR, Pauli E, Kobal G (1997) ‘SniYn’ sticks’: olfactory performance assessed by the combined testing of odor identiWcation, odor discrimination and olfactory threshold. Chem Senses 22:39–52 Hummel T, Konnerth OG, Rosenheim K, Kobal G (2001) Screening of olfactory function with a four-minute odor identiWcation test: reliability, normative data, and investigations in patients with olfactory loss. Ann Otol Rhinol Laryngol 110:976–981 Hygge S (2003) Classroom experiments on the eVects of diVerent noise sources and sound levels on long-term recall and recognition in children. Appl Cogn Psychol 17:895–914 Jäncke L, Sandmann P (2010) Music listening while you learn: no inXuence of background music on verbal learning. Behav Brain Funct 6:3 Larsson M, Nilsson LG, Olofsson JK, Nordin S (2004) Demographic and cognitive predictors of cued odor identiWcation: evidence from a population-based study. Chem Senses 29:547–554 LeCompte DC, Neely CB, Wilson JR (1997) Irrelevant speech and irrelevant tones: the relative importance of speech to the irrelevant speech eVect. J Exp Psychol Learn Mem Cogn 23:472–483 Mozolic J, Joyner D, Hugenschmidt CE, PeiVer AM, Kraft RA, Maldjian JA, Laurienti PJ (2008) Cross-modal deactivations during modality-speciWc selective attention. BMC Neurol 8:35 Nantais KM, Schellenberg EG (1999) The Mozart eVect: an artifact of preference. Psychol Sci 10:370–373 OldWeld RC (1971) The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9:97–113 Oldham GR, Cummings A, Mischel LJ, Schmidtke JM, Zhou J (1995) Listen while you work? Quasi-experimental relations between personal-stereo headset use and employee work responses. J Appl Psychol 80:547–564 Pietschnig J, Voracek M, Formann AK (2010) Mozart eVect-Shmozart eVect: a meta-analysis. Intelligence 38:314–323 Plailly J, Tillmann B, Royet J-P (2007) The feeling of familiarity of music and odors: the same neural signature? Cereb Cortex 17:2650–2658 Rauscher FH, Shaw GL, Ky KN (1993) Music and spatial task performance. Nature 365:611 Rolls ET, Grabenhorst F (2008) The orbitofrontal cortex and beyond: from aVect to decision-making. Prog Neurobiol 86:216–244

123

Exp Brain Res (2011) 212:305–314 Royet J-P, Plailly J, Delon-Martin C, Kareken DA, Segebarth C (2003) fMRI of emotional responses to odors: inXuence of hedonic valence and judgment, handedness, and gender. Neuroimage 20:713–728 Seo H-S, Hummel T (2011) Auditory-olfactory integration: congruent or pleasant sounds amplify odor pleasantness. Chem Senses 36:301–309 Seo H-S, Jeon KJ, Hummel T, Min BC (2009) InXuences of olfactory impairment on depression, cognitive performane, and quality of life in Korean elderly. Eur Arch Otorhinolaryngol 266:1739– 1745 Smith WA (1961) EVects of industrial music in a work situation requiring complex mental activity. Psychol Rep 8:159–162 Spangenberg ER, Grohmann B, Sprott DE (2005) It’s a beginning to smell (and sound) a lot like Christmas: the interactive eVects of ambient scent and music in a retail setting. J Bus Res 58:1583– 1589 Steele KM, Bass KE, Crook MD (1999) The mystery of the Mozart eVect: failure to replicate. Psychol Sci 10:366–369 Thompson WF, Schellenberg EG, Husain G (2001) Arousal, mood, and the Mozart eVect. Psychol Sci 12:248–251 Trimmel M, Poelzl G (2006) Impact of backgrond noise on reaction time and brain DC potential changes of VDT-based spatial attention. Ergonomics 49:202–208 Turner ML, Fernandez JE, Nelson K (1996) The eVect of music amplitude on the reaction to unexpected visual events. J Gen Psychol 123:51–62 Van Gerven PW, Meijer WA, Vermeeren A, Vuurman EF, Jolles J (2007) The irrelevant speech eVect and the level of interference in aging. Exp Aging Res 33:323–339 Vickers ZM, Bourne MC (1976) A psychoacoustical theory of crispness. J Food Sci 41:1158–1164 Werner-Reiss U, Kelly KA, Trause AS, Underhill AM, Groh JM (2003) Eye position aVects activity in primary auditory cortex of primates. Curr Biol 13:554–562 Wesson DW, Wilson DA (2010) Smelling sounds: olfactory-auditory sensory convergence in the olfactory tubercle. J Neurosci 30:3013–3021 Wijayanto T, Tochihara Y, Wijaya AR, Hermawati S (2009) Combined factors eVect of menstrual cycle and background noise on visual inspection task performance: a simulation-based task. J Physiol Anthropol 28:253–259 Woods AT, PoliakoV E, Lloyd DM, Kuenzel J, Hodson R, Gonda H, Batchelor J, Dijksterhuis GB, Thomas A (2011) EVect of background noise on food perception. Food Qual Pref 22:42–47 Zampini M, Spence C (2004) The role of auditory cues in modulating the perceived crispness and staleness of potato chips. J Sens Stud 19:347–363