Abstract
Prior research shows that work on board vessels of the Royal Norwegian Navy (RNoN) is associated with noise exposure levels above recommended standards. Further, noise exposure has been found to impair cognitive performance in environmental, occupational, and experimental settings, although prior research in naval and maritime settings is sparse. The aim of this study was to evaluate cognitive performance after exposure to noise among personnel working on board vessels in the RNoN. Altogether 87 Navy personnel (80 men, 7 women; 31 ± 9 years) from 24 RNoN vessels were included. Noise exposure was recorded by personal noise dosimeters at a minimum of 4 h prior to testing, and categorized into 4 groups for the analysis: <72.6 dB(A), 72.6-77.0 dB(A), 77.1-85.2 dB(A), and >85.2 dB(A). The participants performed a visual attention test based on the Posner cue-target paradigm. Multivariable general linear model (GLM) analyses were performed to analyze whether noise exposure was associated with response time (RT) when adjusting for the covariates age, alertness, workload, noise exposure in test location, sleep the night before testing, use of hearing protection device (HPD), and percentage of errors. When adjusting for covariates, RT was significantly increased among personnel exposed to >85.2 dB(A) and 77.1-85.2 dB(A) compared to personnel exposed to <72.6 dB(A).
Keywords: Cognitive performance, Navy, noise exposure, Posner cue-target paradigm, vessels, visual attention
Introduction
Prior research has been conducted to assess the relationship between noise exposure and cognitive performance;[1] however, the results have been ambiguous as the effects of noise on performance have been found to be facilitative, detrimental, or even absent.[2] A recent literature review summarizes that noise exposure does have deleterious effects on cognitive performance but that the magnitude of these effects is dependent on various factors such as noise intensity, noise duration, and task performance.[3]
Measurements on board Royal Norwegian Navy (RNoN) vessels reveal noise levels ranging from 45 to 75 dB(A) depending on vessel class, and in many vessels noise levels exceed recommended standards.[4,5] Prior research has found noise exposure to increase the risk of occupational accidents.[6,7,8] The RNoN has experienced a number of navigation accidents at sea,[9] and it is reasonable to question whether such accidents can be influenced by failure in performance due to noise exposure.
Quite a few maritime field studies address the effects of various stressors on seafarers’ health.[10,11,12] However, research on noise exposure and cognitive performance in maritime and naval settings is sparse.[10,13,14,15] The European Network on Noise and Health has recommended further research on acute and chronic exposure to levels of noise less than 70 dB(A) as well as examination of the exposure-effect relationship of noise on cognition in different contexts.[16] We aimed to assess cognitive performance after noise exposure among Navy personnel on board RNoN vessels using personal noise dosimeter measurements and a test of visual attention. We hypothesized impaired cognitive performance after exposure to noise.
Methods
This study is a part of the RNoN project “Noise and Health in the Navy.”
Data were collected on board 24 Navy vessels by two trained university researchers from April 2012 to June 2013. The study was approved by the Regional Committee of Medical and Health Research Ethics (REC South East) and by the RNoN. The results were treated anonymously and no individual data were accessible to the RNoN. The participants were informed about the study and signed their informed consent. The participants could withdraw from the study at any point.
Participants
On board each vessel, the management was requested to identify two to five potential participants with different levels of anticipated noise exposure on board. These persons were then informed about the study and the researchers invited them to participate. Only one of the invited participants did not consent to participate. A total of 116 healthy Navy personnel (response rate of 99%), including officers and enlisted personnel, participated in the study. All Navy personnel have to fulfil strict health requirements (physical and mental) in order to work on board RNoN vessels, thus the general health status in this population was considered to be very good. The participants had Norwegian as their native language.
Design
All participants had their cognitive performance assessed by a visual attention test twice on the same day, with a minimum of 4 h in between (7.5 ± 2.5 h, range: 4.3-9.5 h). The first test served as a learning session. Individual noise exposure was recorded between the two tests for all participants. Participants also each kept a log book containing data about time spent in different locations, sleep the night before testing, and various other factors that possibly could influence performance. The log books were completed under guidance from the researchers.
Cognitive performance test
The cognitive performance test was based on the Posner cue-target paradigm, a test that assesses response time (RT), accuracy, and inhibition,[17,18] and was programmed using E-Prime 2.0, standard version (Psychology Software Tools, Inc., Sharpsburg, PA, USA).
The test was conducted twice, the learning session in the morning and the second test in the afternoon. However, in order to comply with work requirements on board, some participants had to complete their learning session in the afternoon and the second test in the evening. The test was presented on a laptop with a 13.3" screen. The screen display consisted of a fixation cross hair in the middle and two horizontal rectangles located on each side of the cross hair [Figure 1].
Figure 1.
An overview of the cognitive performance test on Navy personnel. (a) Screen display with fixation cross hair and horizontal rectangles (b) Stimulus presentation without cue (no cue) (c) Stimulus presentation following a valid cue (d) Stimulus presentation following an invalid cue
The participants were instructed to fixate on the cross hair and to respond as fast as possible with their left index finger by pressing “d” on the laptop keyboard when the target stimulus (an asterisk) appeared in the left rectangle, or by pressing “l” on the keyboard with the right index finger when the target stimulus appeared in the right rectangle (these two letters were seen as optimal according to their placement on the keyboard). Before the target stimulus appeared, the frame of one of the rectangles sometimes became broader (i.e., a cue). The participants were informed to ignore this cue stimulus and only press the keyboard when the asterisk appeared.
When the frame did not become broader in advance of the target stimulus, it was called a no cue presentation [Figure 1b]. When the target stimulus appeared inside the rectangle with a broader frame, this was called a valid cue presentation (i.e., the cue and the target stimulus had the same spatial location) [Figure 1c]. A valid cue presentation would cause a shift in attention toward the rectangle with the target stimulus, hence decreasing RT. When the target stimulus appeared inside the rectangle opposite to the cue location, it was called an invalid cue presentation [Figure 1d]. This presentation would misdirect attention to the opposite rectangle, hence increasing RT. All participants were instructed orally. The entire test lasted 4.40 min.
In total, 168 target stimuli were presented during one test session. Each target stimulus was presented for 500 milliseconds (ms) with interstimuli intervals of 600-1400 ms. The cue stimulus was presented for 200 ms or 400 ms before the target stimulus appeared. Twenty-eight (16.7%) of the target stimuli appeared without cue (i.e., no cue), 112 (66.7%) of the target stimuli appeared in the same rectangle as the cue (i.e., valid cue), and 28 (16.7%) of the target stimuli appeared in the opposite rectangle to the cue (i.e., invalid cue).
RT and percentage of errors (i.e., response accuracy) were recorded and stored on the laptop for each trial. Responses before the target stimulus appeared and responses during the first 99 ms after target presentation were defined as erroneous responses. In addition, responses recorded with the wrong response button (for example, “l” instead of “d”) were also considered erroneous responses. If the participants corrected an erroneous response by pressing a second time before the next stimulus was presented, the response was considered a correct response.
The cognitive performance tests were performed in a preferably quiet and undisturbed location (usually the sick bay or a cabin) on board each vessel. Selection of location was made in cooperation with the management. The locations were equipped with a desk onto which the laptop was placed. Background noise levels in these test locations were measured with noise dosimeters during the second test.
The Posner cue-target paradigm was selected as a suitable test to assess cognitive performance among Navy personnel as it reflects the ability to respond quickly (RT) and correctly (response accuracy) to a stimulus and the ability to inhibit distracting cues (response inhibition), all important factors to secure efficiency and safety on board Navy vessels.
Noise dosimeters
Personal noise dosimeters (Brüel and Kjaer Type 4445 or 4448, Nærum, Denmark) were used to assess individual noise exposure during work before the cognitive performance test. The noise dosimeter was mounted on the right shoulder of the participants after the first test and recorded equivalent noise level (in LAeq) each minute until the completion of the second test. Noise measurements were processed using the Brüel and Kjaer program, Protector Type 7825 Version 5.0.0 (Nærum, Denmark), where equivalent noise levels for the required time periods (i.e., last 4 h before and 5 min periods during the second test) was automatically calculated. For eight participants, the noise dosimeter recordings were stopped just before the second test was conducted. As the test locations were the same for participants on the same vessel, the missing data were replaced with the mean noise level during the second test for the other participants on the same vessel. The participants were informed that contact with the microphone during noise measurements should be avoided in order to reduce disturbance. If putting on any additional garment, they were asked to place the noise dosimeter on top of it. The noise dosimeter's measurement range was 50-120 dB(A) for Type 4445 and 50-140 dB(A) for Type 4448. The time weighting was set to “Fast” and a 3 dB exchange rate was used. The noise dosimeters were calibrated at the beginning and at the end of each measurement period. No significant shift in calibration was detected.
Log books
Printed log books were handed out during the evening before the first test and were completed by the participants in writing between tests and handed in by the end of the second test. If the participants had any questions, they could consult the researchers whenever they wished throughout the day. In order to assess compliance, the researchers went through the log book together with each participant after completion of the second cognitive performance test. In the log book, they filled in the time spent in different locations on board and specified when and for how long they had slept. We chose to categorize the amount of sleep the night before testing into two categories: Having had at least 6 h of continuous sleep from 1 AM or earlier the previous night, or sleep deviating from the criteria mentioned. If data on sleep were missing for any participants or participants reported sleep between the two cognitive performance tests, they were excluded. All participants indicated their alertness, i.e., how well-disposed they felt before the test by placing a mark on a 10-cm horizontal line toward either extreme of the scale (not disposed/very disposed). This method was also used to assess workload during the last work shift. In addition, use of caffeine (number of cups of coffee), nicotine (number of cigarettes or portions of moist snuff), medication (permanent/temporary medication), and hearing protection device (HPD) between tests was noted in the log book.
Data analysis
To investigate the relationship between noise exposure and cognitive performance, the equivalent noise levels measured the last 4 h before the second test were grouped by quartiles, <72.6 dB(A), 72.6-77.0 dB(A), 77.1-85.2 dB(A), and >85.2 dB(A). We chose to group by quartiles as we could not any find cut-off values for noise exposure levels related to cognitive function in the literature. In order to evaluate cognitive performance after exposure to noise, only results from the second cognitive performance test were analyzed. Regarding cognitive performance (RT) and percentage of errors, all statistical analyses were performed separately for no cue-, valid cue-, and invalid cue-stimuli presentations. The covariates applied were age, gender, self-reported alertness (0-10), self-reported workload (0-10), noise exposure in test location, sleep the night before testing (≥6 h/<6 h), coffee intake (yes/no), use of nicotine (yes/no), and use of HPD the last 4 h (yes/no). Descriptive statistics for the covariates and noise exposure levels were provided with mean values, standard deviation (SD), standard error of the mean (SEM), and percentages.
Separate analyses were made in order to find covariates associated with noise exposure or covariates associated with RT. Associations between the single covariates and the noise exposure groups were tested with analysis of variance (ANOVA), the chi-square test, and Fisher's exact test (if there were few numbers in the exposure groups). Linear regression analysis was used to identify associations between the single covariates and RT.
Mean and SD for RT and percentage of errors were calculated for the noise exposure groups, and associations were tested by ANOVA.
Estimated mean and SEM for RT adjusted for percentage of errors was calculated by ANOVA to control for speed-accuracy trade-off effects.
Finally, multivariable general linear model (GLM) analyses were used to investigate the association between noise exposure and RT. Covariates that differed significantly between the noise exposure levels or were associated with RT with a P value < 0.05 for at least one of three stimuli presentations were included in the final models: Age, alertness, workload, noise exposure in test location, sleep the night before testing, and use of HPD. Additionally, percentage of errors was included in order to adjust for speed-accuracy trade-off effects. The multivariable GLM analyses estimated adjusted differences in RT means [with noise exposure level <72.6 dB(A) as a reference] with a 95% confidence interval (CI). The Statistical Products of Service Solution package (IBM SPSS Statistics, version 22, Armonk, NY, USA) was used for all statistical analyses. P values < 0.05 were considered to be statistically significant.
Results
Among the 116 healthy Navy personnel, altogether 29 participants were excluded due to missing data, and the analyzed material consisted of 87 participants: 80 men and 7 women aged 18-61 years (31 ± 9 years). Individual noise exposure in the 4-h period prior to the second test varied from 67.2 to 99.1 dB(A), with a median level at 77.0 dB(A). The participants with the lowest noise exposure, <72.6 dB(A), expressed the highest grade of subjective alertness; however, there was no significant difference in subjective alertness between the noise exposure groups [Table 1]. Participants with noise exposure <72.6 dB(A) reported the lowest workload, and there was a significant difference between grades of claimed workload among the noise exposure groups. When comparing mean noise exposure in test locations throughout the 4 noise exposure groups, no significant difference was found. There was no significant difference in sleep the night before testing between the noise exposure groups. The consumption of caffeine and the use of nicotine did not differ significantly between the noise exposure groups. Participants with the highest noise exposure had the highest prevalence of HPD use [Table 1]. There was a significant difference in use of HPD between the noise exposure groups. None of the log books contained information about use of any medication known to affect performance.
Table 1.
Characteristics of personnel on board Navy vessels by equivalent noise exposure levels (in quartiles) measured before the cognitive performance test
| Covariate | Equivalent noise exposure levels [dB(A)] measured before the cognitive performance test | P-value | |||
|---|---|---|---|---|---|
| <72.6 | 72.6-77.0 | 77.1-85.2 | >85.2 | ||
| Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | ||
| Age | 33 (9) | 29 (8) | 28 (7) | 33 (11) | 0.178a |
| Alertness1 | 7.2 (1.6) | 6.8 (2.1) | 6.2 (1.9) | 6.8 (2.1) | 0.443a |
| Workload2 | 2.9 (1.6) | 4.7 (1.9) | 4.5 (1.9) | 4.0 (2.0) | 0.010a |
| Mean noise exposure level [dB(A)] in test location | 58 (4) | 60 (6) | 59 (4) | 60 (5) | 0.664a |
| n (%) | n (%) | n (%) | n (%) | ||
| Gender | |||||
| Male | 21 (95) | 18 (82) | 20 (91) | 21 (100) | 0.201b |
| Female | 1 (5) | 4 (18) | 2 (9) | 0 | |
| Sleep, night before testing | |||||
| ≥6 h of continuous sleep | 12 (55) | 18 (82) | 13 (59) | 17 (81) | 0.100c |
| <6 h of continuous sleep | 10 (45) | 4 (18) | 9 (41) | 4 (19) | |
| Coffee | |||||
| Yes | 15 (68) | 14 (64) | 17 (77) | 15 (71) | 0.793c |
| No | 7 (32) | 8 (36) | 5 (23) | 6 (29) | |
| Nicotine | |||||
| Yes | 9 (41) | 8 (36) | 12 (55) | 12 (57) | 0.441c |
| No | 13 (59) | 14 (64) | 10 (45) | 9 (43) | |
| HPD | |||||
| Yes | 0 (0) | 3 (14) | 7 (32) | 15 (71) | <0.001c |
| No | 22 (100) | 19 (86) | 15 (68) | 6 (29) | |
1Marking on 10-cm horizontal line indicating grade of alertness. Missing: n = 4, 2Marking on 10-cm horizontal line indicating grade of workload. Missing: n = 5, aANOVA, bFisher’s exact test, cChi-square test
Linear regression analyses showed a significant association between RT and the covariates age (P < 0.001 for all stimuli presentations), alertness (P = 0.02 for no cue-stimuli presentations), noise exposure in test location (P = 0.02 for no cue-stimuli presentations, P = 0.04 for valid cue-stimuli presentations), and sleep the night before testing (P < 0.05 for all stimuli presentations). There was no association between RT and gender, workload, use of caffeine, nicotine, or HPD, respectively.
RT was the longest for no cue-stimuli presentations, intermediate for invalid cue-stimuli presentations, and shortest for valid cue-stimuli presentations, as observed in all noise exposure groups [Table 2]. A slight increase in RT was found by increasing levels of noise exposure; however, this was not significant. There was no significant association between noise exposure and percentage of errors. The highest percentage of errors was found for invalid cue presentations for all noise exposure groups.
Table 2.
Mean response time (RT) in ms and percentage of errors for different stimuli presentations (no cue, valid cue, and invalid cue) for personnel on board Navy vessels after exposure to different equivalent noise levels (in quartiles)
| Equivalent noise exposure levels [dB(A)] measured before the cognitive performance test | |||||
|---|---|---|---|---|---|
| <72.6 | 72.6-77.0 | 77.1-85.2 | >85.2 | P valuea | |
| Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | ||
| RT (ms)No cue | 371 (36) | 372 (35) | 377 (34) | 380 (40) | 0.84 |
| RT (ms)Valid cue | 285 (33) | 289 (30) | 294 (28) | 298 (39) | 0.57 |
| RT (ms)Invalid cue | 357 (40) | 349 (32) | 361 (31) | 362 (42) | 0.65 |
| Errors (%)No cue | 2.4 (3.3) | 1.9 (4.1) | 2.2 (3.3) | 2.0 (3.0) | 0.97 |
| Errors (%)Valid cues | 3.7 (4.9) | 2.7 (2.5) | 2.7 (2.8) | 2.7 (2.1) | 0.71 |
| Errors (%)Invalid cues | 8.5 (10.1) | 3.6 (4.2) | 6.1 (5.9) | 5.1 (4.5) | 0.10 |
aANOVA
We found longer RTs for no cue- and valid cue-stimuli presentations by increasing noise exposure levels when adjusting for percentage of errors, although differences were small [Figure 2].
Figure 2.
Mean response time (RT) adjusted for percentage of errors and presented with SEM, for different stimuli presentations (no cue, valid cue, and invalid cue) among personnel on board Navy vessels exposed to different equivalent noise levels (in quartiles) measured before the cognitive performance test
When adjusting for age, alertness, work load, noise exposure in test location, sleep the night before testing, HPD, and percentage of errors, a significant increase in RT was found among personnel working in the > 85.2 dB(A) noise exposure group compared to the reference [<72.6 dB(A)] for no cue-stimuli presentations [Table 3]. A significant increase in RTs was found when comparing the 77.1-85.2 dB(A) noise exposure group with the reference [<72.6 dB(A)] for no cue- and valid cue-stimuli presentations.
Table 3.
Difference in mean adjusteda response time (RT) in ms for different stimuli presentations (no cue, valid cue, and invalid cue) for personnel on board Navy vessels after exposure to different equivalent noise levels (in quartiles) compared to the reference group [<72.6 dB(A)]
| Equivalent noise exposure levels [dB(A)] measured before the cognitive performance test | |||||||
|---|---|---|---|---|---|---|---|
| Stimuli presentations | <72.6 | 72.6-77.0 | 77.1-85.2 | >85.2 | |||
| RT (ms) | [95% CI] | RT (ms) | [95% CI] | RT (ms) | [95% CI] | ||
| No cue | ref | 8.7 | [-8.3-25.8] | 23.8 | [6.4-41.3]* | 24.6 | [5.0-44.2]* |
| Valid cue | ref | 10.1 | [-6.8-26.9] | 25.0 | [7.7-42.4]* | 18.9 | [-0.6-38.4] |
| Invalid cue | ref | -1.5 | [-22.6-19.6] | 18.7 | [-2.6-40.1] | 12.9 | [-11.0-36.8] |
aAdjusted for age, alertness, workload, noise exposure in test location, sleep the night before testing, HPD use, and percentage of errors, *P value < 0.05
Discussion
The present study found significantly increased RT among participants with the highest noise exposure [>85.2 dB(A)] (no cue presentations) and the next highest exposure [77.1-85.2 dB(A)] (no cue and valid cue presentations) compared to participants with the lowest noise exposure [<72.6 dB(A)]. However, there was a lack of difference between the highest and lowest noise exposure groups for valid and invalid cue presentations. This might be explained by more extensive use of HPD among the highest-exposed personnel: Not being as highly exposed as the measured noise level indicates. A method to assess individual noise reduction by use of HPD has been suggested in a prior RNoN study.[5] However, we had limited individual information about HPD use (selection and fitting of earplugs and/or muffs and duration of use), and due to these uncertainties we could not make a detailed evaluation of use of HPD. This made our adjustment for HPD in the analysis likely to have major weaknesses, disturbing a possible dose-response effect from higher noise levels in our material.
Investigation of cognitive performance after exposure to noise has received relatively little attention in prior research.[19,20,21,22,23] Several observational and experimental studies have investigated the acute effects of noise exposure during completion of different cognitive tasks, but these findings are of limited interest as acute effects during exposure are not required to recognize subsequent effects.[24]
Prior research on noise exposure and cognitive performance in maritime and naval settings is limited. We are aware of only one study conducted in the Navy in which noise exposure levels and cognitive performance were evaluated.[15] In this study, noise exposure levels were measured on board three ships in the coastal fleet (two patrol boats and one experimental ship). Simple reaction time was measured in the morning, at noon, and in the afternoon among the 29 participants working on board. Performance at sea [about 80 dB(A)] was compared to performance at quay [50-65 dB(A)]. In the experimental ship, prolonged RT was observed in the afternoon and this effect was more pronounced at sea. This finding supports our suggestion of a relationship between low performance and high noise exposure. However, opposite results were found for the patrol boats. Here RT was shorter in the afternoon than in the morning (as expected according to diurnal rhythm) and there were no differences in RT at sea or at quay. As the noise exposure levels on board the two types of ships were similar, it was argued that the impaired performance among those working on board the experimental ship might be due to the low-frequency character of the noise on board.
Noise exposure and cognitive function have also been studied in a research program on seafarer's fatigue. In the first and second phases of the program, seafarers working in the offshore oil industry and the short sea sector were studied.[25] Noise exposure was measured using noise dosimeters and cognitive function was evaluated through simple reaction time, a focused attention task, and a categorical search task before and after work shifts. In the offshore oil industry, 62 participants on board three vessels were examined. Postshift mean simple RT showed that those exposed to ≥59 dB(A) had a significantly longer RT (314.4 ms) than those exposed to <59 dB(A) (288.6 ms). Significant differences between the two noise exposure groups were also seen for the focused attention test and the categorical search task. This is in line with the findings in our present study. However, the significant findings from the study conducted in the offshore oil industry might be explained by confounding factors, as discussed by the authors.[13] In the short sea shipping industry study, data were collected from 177 participants working on board seven vessels. A significant correlation between noise levels and increased ability to encode new information on a focused attention task was found. However, there were no significant differences seen for the other performance measures in the shipping industry.[14] It was suggested that the inconsistent findings of the studies from the research program were related to different noise characteristics on board the vessels. Prior studies have suggested intermittent noise to be more disruptive than continuous noise.[3,24,26,27] The noise exposure on board the vessels serving the offshore oil industry was intermittent, as alarms and closing doors were common, while noise in the short sea shipping industry study was more continuous (e.g., background noise from engines). In our study the greater part of noise exposure was continuous; however, there were also some intermittent elements.
The methodology used to assess cognitive performance varies extensively. As vigilance has been found to be particularly affected by noise,[2] we chose to assess the relationship between noise exposure and cognitive performance using the Posner cue-target paradigm. The test is not time-consuming and is simple to carry out on board. It also indicates the ability to maintain concentration and apprehend significant signals and events of importance: Skills that are crucial when working on board Navy vessels. Prior research has stated that speed of performance is not influenced by noise exposure, contrary to performance related to accuracy.[3] This is not supported by our findings, as we found longer RT by increasing noise exposure and no relationship between noise exposure and percentage of errors. The number of erroneous responses has been found to increase by increasing task duration,[24] thus the missing association found in our study might be explained by low task duration. It has been generalized that complex task performance is more likely to be impaired by noise, while simple task performance can be improved.[2] Consequently, selecting a more complex task might have challenged the participant to a greater extent, and the impact of noise exposure could then have been more perceptible.
Personal characteristics might influence cognitive performance when an individual is exposed to noise;[27,28] some individuals are disturbed, others are not affected, while some can improve performance when being noise-exposed.[2] The impact of different personal characteristics may be of less importance in this study as the Navy personnel are selected for work on board by specific criteria, limiting the probability of mental illness and personality disorders. It is likely that this group is healthier than civilians of similar gender and age, and therefore may show enhanced performance. Contradictory to what we expected, mean RT was shorter for invalid cue presentations than for no cue presentations, hence cue inhibition was not affected. However, the percentage of errors was higher for invalid cue presentations than for the other presentations. This speed-accuracy trade-off effect might be explained by presumably competitive participants.
A complete recording of individual, organizational, and incidental factors with possible impact on cognitive performance is not feasible in maritime field studies. A strength of our study was the use of personal noise dosimeters, as these probably reflect individual noise exposure even more accurately than area measurements, and thus we were able to group the participants according to a valid noise exposure. In prior maritime research,[13,14] noise dosimeters have been placed on the bridge and in a cabin on board the vessels, thus giving a poorer indication of individual noise exposure. Previous research has indicated impaired performance by exposure to low-frequency noise.[15] However, the noise dosimeters used in our study did not provide frequency analysis, making it difficult to evaluate the influence of frequency composition. The effect of noise has been found to vary over time, perhaps as a result of adaptation, as short-duration exposure has been shown to be more detrimental to performance than long duration.[3,24] Unfortunately, we had limited information about time spent on board prior to data collection, and changes in RT over time were not measured; however, we do not know if more information of this kind would have influenced the results. We did not find any relationship between caffeine intake and cognitive performance. However, we did not collect information about caffeine habits such as tolerance and abstinence, which might have an impact on cognitive performance.[29] Another strength of the study was the high response rate: All but one person invited participated. Although we had relatively few participants in each noise exposure group, thus providing a rather low statistical power, we were able to find statistically significant associations between noise exposure and cognitive performance. However, we cannot exclude the possibility of more significant findings with an increased number of participants. Our aim was to assess cognitive performance after exposure to noise. However, our results might also have been influenced by acute effects of noise exposure, as we did not provide HPDs to be worn during testing. Still, the contribution of acute effects was probably limited, as noise exposure in the test locations was equally distributed across the noise exposure groups.
As prior research is limited, cognitive performance succeeding noise exposure in maritime settings should be assessed further. We suggest that future studies compare results from cognitive performance tests conducted prior to and after noise exposure and put emphasis on detailed data registration (e.g., personal characteristics, caffeine habits, and HPD use).
The results from this study state that high noise exposure levels during work on board Navy vessels impair RT, as assessed by a test of visual attention. We found no association between noise exposure and test errors (response accuracy), and the ability to inhibit invalid cue presentations was not affected (response inhibition); however, selecting a more cognitively challenging test with longer test duration might have changed this outcome. Our findings have implications for work on board Navy vessels as quick and correct responses are important in military operations (e.g., shooting and navigation). Noise-reducing (e.g., attenuation of noise sources, correct and increased use of HPD) and performance-shaping measures (e.g., training, task organization, and improved bridge systems)[9] should be considered in order to improve RT among Navy personnel on board RNoN vessels.
Financial support and sponsorship
The study was funded by the Royal Norwegian Navy.
Conflicts of interest
There are no conflicts of interest.
Acknowledgment
We thank the RNoN for their cooperation and wish to show our appreciation to the Navy personnel who participated in the project, the management on board the vessels, and Hjalmar Johansen and Vilhelm Koefoed for being contact persons in the RNoN, Gunhild Koldal for registering the data, Lorentz Irgens for valuable contributions in the writing process, Ågot Irgens for helpful support with the statistics, Jude Nicholas for expertise on neuropsychology, Gunnhild Oftedal, Truls Gjestland and Ole Jacob Møllerløkken for inspiring ideas, and Camilla Hauge for collecting and registering data.
References
- 1.Basner M, Babisch W, Davis A, Brink M, Clark C, Janssen S, et al. Auditory and non-auditory effects of noise on health. Lancet. 2014;383:1325–32. doi: 10.1016/S0140-6736(13)61613-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Koelega HS, Brinkman JA. Noise and vigilance: An evaluative review. Hum Factors. 1986;28:465–81. doi: 10.1177/001872088602800408. [DOI] [PubMed] [Google Scholar]
- 3.Szalma JL, Hancock PA. Noise effects on human performance: A meta-analytic synthesis. Psychol Bull. 2011;137:682–707. doi: 10.1037/a0023987. [DOI] [PubMed] [Google Scholar]
- 4.Haakonsvern: The Royal Norwegian Navy Standard Requirements and Regulations; 2008. Norwegian Defence Systems Management Division; pp. 4–8. [Google Scholar]
- 5.Sunde E, Irgens-Hansen K, Moen BE, Gjestland T, Koefoed VF, Oftedal G, et al. Noise and exposure of personnel aboard vessels in the Royal Norwegian Navy. Ann Occup Hyg. 2015;59:182–99. doi: 10.1093/annhyg/meu075. [DOI] [PubMed] [Google Scholar]
- 6.Smith A. Noise, performance efficiency and safety. Int Arch Occup Environ Health. 1990;62:1–5. doi: 10.1007/BF00397841. [DOI] [PubMed] [Google Scholar]
- 7.Toppila E, Pyykkö I, Pääkkönen R. Evaluation of the increased accident risk from workplace noise. Int J Occup Saf Ergon. 2009;15:155–62. doi: 10.1080/10803548.2009.11076796. [DOI] [PubMed] [Google Scholar]
- 8.Girard SA, Picard M, Davis AC, Simard M, Larocque R, Leroux T, et al. Multiple work-related accidents: Tracing the role of hearing status and noise exposure. Occup Environ Med. 2009;66:319–24. doi: 10.1136/oem.2007.037713. [DOI] [PubMed] [Google Scholar]
- 9.Gould KS, Røed BK, Koefoed VF, Bridger RS, Moen BE. Performance-shaping factors associated with navigation accidents in the Royal Norwegian Navy. Mil Psychol. 2006;18(Suppl):S111–29. [Google Scholar]
- 10.Oldenburg M, Hogan B, Jensen HJ. Systematic review of maritime field studies about stress and strain in seafaring. Int Arch Occup Environ Health. 2013;86:1–15. doi: 10.1007/s00420-012-0801-5. [DOI] [PubMed] [Google Scholar]
- 11.Oldenburg M, Jensen HJ, Latza U, Baur X. Seafaring stressors aboard merchant and passenger ships. Int J Public Health. 2009;54:96–105. doi: 10.1007/s00038-009-7067-z. [DOI] [PubMed] [Google Scholar]
- 12.Allen P, Wadsworth E, Smith A. Seafarers’ fatigue: A review of the recent literature. Int Marit Health. 2008;59:81–92. [PubMed] [Google Scholar]
- 13.Wellens BT, McNamara RL, Ellis N, Smith AP. Combined effects of shift work and occupational noise exposure on performance tasks in a seafaring population. Arch Complex Environ Stud. 2002;14:3–4. [Google Scholar]
- 14.Ellis N, Allen P, Burke A. The influence of noise and motion on sleep, mood, and performance of seafarers. In: McCabe PT, editor. Contemporary Ergonomics. 1st ed. London: Taylor & Francis; 2003. pp. 131–6. [Google Scholar]
- 15.Kjellberg A, Muhr P, Sköldström B. Fatigue after work in noise - An epidemiological survey study and three quasi-experimental field studies. Noise Health. 1998;1:47–55. [PubMed] [Google Scholar]
- 16.Luxembourg: European Network on Noise and Health (ENNAH); 2013. JRC Scientific and Policy Reports. Final Report of the ENNAH (European Network on Noise and Health) Project; p. 102. [Google Scholar]
- 17.Gundersen H, Helland CA, Raeder MB, Hugdahl K, Wester K. Visual attention in patients with intracranial arachnoid cysts. J Neurol. 2007;254:60–6. doi: 10.1007/s00415-006-0280-2. [DOI] [PubMed] [Google Scholar]
- 18.Posner MI. Orienting of attention: Then and now. Q J Exp Psychol (Hove) 2014:1–12. doi: 10.1080/17470218.2014.937446. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ljungberg JK, Parmentier FBR. Psychological effects of combined noise and whole-body vibration: A review and avenues for future research. P I Mech Eng D-J Aut. 2010;224:1289–302. [Google Scholar]
- 20.Cohen S. Aftereffects of stress on human performance and social behavior: A review of research and theory. Psychol Bull. 1980;88:82–108. [PubMed] [Google Scholar]
- 21.Evans GW, Johnson D. Stress and open-office noise. J Appl Psychol. 2000;85:779–83. doi: 10.1037/0021-9010.85.5.779. [DOI] [PubMed] [Google Scholar]
- 22.Ljungberg JK, Neely G. Cognitive after-effects of vibration and noise exposure and the role of subjective noise sensitivity. J Occup Health. 2007;49:111–6. doi: 10.1539/joh.49.111. [DOI] [PubMed] [Google Scholar]
- 23.Glass DC, Singer JE. Behavioral aftereffects of unpredictable and uncontrollable aversive events. Am Sci. 1972;60:457–65. [PubMed] [Google Scholar]
- 24.Kjellberg A. Subjective, behavioral and psychophysiological effects of noise. Scand J Work Environ Health. 1990;16(Suppl 1):29–38. doi: 10.5271/sjweh.1825. [DOI] [PubMed] [Google Scholar]
- 25.Smith A, Allen P, Wadsworth E. Seafarers Fatigue: The Cardiff Research Programme. Cardiff: Centre for Occupational and Health Psychology, Cardiff University; 2006. [Google Scholar]
- 26.Smith A. An update on noise and performance: Comment on Szalma and Hancock (2011) Psychol Bull. 2012;138:1262–73. doi: 10.1037/a0028867. [DOI] [PubMed] [Google Scholar]
- 27.Smith A. A review of the effects of noise on human performance. Scand J Psychol. 1989;30:185–206. doi: 10.1111/j.1467-9450.1989.tb01082.x. [DOI] [PubMed] [Google Scholar]
- 28.Belojevic G, Slepcevic V, Jakovljevic B. Mental performance in noise: The role of introversion. J Environ Psychol. 2001;21:209–13. [Google Scholar]
- 29.Rogers PJ, Heatherley SV, Mullings EL, Smith JE. Faster but not smarter: Effects of caffeine and caffeine withdrawal on alertness and performance. Psychopharmacology (Berl) 2013;226:229–40. doi: 10.1007/s00213-012-2889-4. [DOI] [PubMed] [Google Scholar]