apcswg april 2012 -molesworth
TRANSCRIPT
5/04/12
1
The Effects of Noise Cancelling Headphones on Performance
Introduction
Brett Molesworth PhD – Senior Lecturer Safety and Human Factors – Commercial Pilots Licence (CPL) – Advanced Aerobatics – Psychologist NSW – PhD Psychology – Human Factors – BAv (Hons) Aviation Human Factors – BAv – Aviation Management, Psychology
Background
Asia Pacific Cabin Safety Working Group
– Passenger uses Active Noise Reduction headphones during pre-flight safety brief
– Legislative concern – PED may interfere with safe operation of aircraft - communication and navigation equipment (CASA AC 91-050(0) 2001)
– Passenger claims – allows him to hear safety brief better (increases signal to noise ratio)
5/04/12
2
Active Noise Reduction ���Technology
Pseudo name – Noise cancelling headphones
In Theory – Produces a sound that is 180
degrees out of phase with the original sound
– Two sounds combined cancels noise (Nelson & Elliott, 1992).
Noise
• Noise defined – Any stimuli that is unwanted
• It may be unwanted because: – Unpleasant – Harmful – Distractor
So what is Noise?
5/04/12
3
So what is Noise?
So what is Noise?
So what is Noise?
5/04/12
4
Audio Noise and��� Performance
• Audio Noise defined - Any sound stimuli that is unwanted (Antunana & Spanyers, 2000)
• Effects: – cause hearing impairment (Daniel, 2007; Cruickshanks et al.,
2010)
– induce stress (Taffinder, McManus, Gul, Russell, & Darzi, 1998)
– cause fatigue (Picard et al., 2008)
– alter health state (Gangwisch et al., 2006)
– negatively impacts memory (Sorqvist, 2010)
– increase error rate (Weinger & Ancoli-Israel, 2002)
The Effects of Noise ���on Performance
• Recognised by World Health Organisation (WHO), Hearing protective required when employees are subjected to occupational noise > 90 dB(A).
• NOHSC - 85dB(A) for no more than 8 hours (NOHSC:
1007, 2000)
• AS/NZS 2107 office noise ~ 40-45 dB(A)
Noise, Cognition and ���Performance
• Stimuli processed – Explicitly, and – Implicitly (LaPointe et al., 2007)
• Explicit = distractor (consumes cognitive resources) • Implicit = consumes limited cognitive resource
5/04/12
5
Noise, Cognition and ���Performance
Study 1
Context • Commercial aviation • Headphones banned during taxi phases of flight • Engines operational • Pre-flight safety brief provided • Noise affects the intelligibility of the message
In other words, • Decreases signal to noise ratio
Study 1
Aim • Examine the effects of noise cancelling headphones on
performance (commercial aviation)
Participants • 25 (12 male) • Average age 26 years • Normal hearing (tested)
5/04/12
6
Study 1
Procedure • Hearing test • 5 audio conditions with 5 audio briefs • Written fill-in-the-blanks test after each condition
• Presented in a Latin square design with audio briefs systematically varied
Study 1
Audio Condition
1 2 3 4 5
NC Passive ✓
NC Active ✓ ✓
✓
Brief through speaker
✓
✓
✓
✓
Brief through headphone
✓
Music ✓
Wideband noise
✓
✓
✓
✓
✓
Study 1- Design
Design • 5 x 5 Latin square design • Order of stimuli systematically varied • DV - # of correct responses (Max 12) • Series of planned comparisons with alpha adjusted
A B C D E B C D E A C D E A B D E A B C E A B C D
5/04/12
7
Study 1- Design
Design • Aircraft noise analysed and recorded - 65 dB(A) • Consistent with Ozcan and Nemlioglu, (2006) • Safety Brief – set by flight attendant with > 16 years • Safety Brief set at 70 dB(A) • Conducted in quiet room 38 dB(A)
Study 1- Results
Audio Condition
Mean Correct Mean % Correct
Range Median
1 4.40 (2.24) 36.67% 0-9 5
2 3.92 (2.33) 33.66% 1-9 4
3 2.84 (1.89)* 23.67% 0-7 3
4 3.00 (2.36)* 25.00% 0-8 2
5 0.68 (1.07)* 5.67% 0-4 0
1-2 4.16 (2.24)* 34.67% 1-8.5 4
Study 1- Results
• Active noise cancelling headphones aided in the recognition of information irrespective of the source of the target audio brief (e.g., through headphones or external speaker).
• Performance deteriorated when: – headphones not used, – the active noise cancelling feature of the headphones was
switched off, or – active noise cancellation with music.
5/04/12
8
Study 1- Conclusion
Active noise cancelling increases: • signal to noise ratio, thereby • increasing intelligibility, • ultimately improving recall.
But what about: • Type of Headphone, • Dual task, • Masking, • Age, and • Native language?
Study 2 – Dual Task
Aim 1. Examine differences in performance between two
commercially available headphones 2. Examine the effects of noise cancelling headphones on
concurrent task performance (commercial aviation)
Participants • 36 (23 male) • Average age 20 years • Normal hearing (tested)
Study 2 – Dual Task
Design • Balanced Latin square design (6 x 6) • DV – correct # on fill-in-the-blanks test (max 12) • Aircraft noise 65 dB(A) • Audio briefs 70 dB(A) • Concurrent task – maths question summing to < 20
– E.g., 3 + 7, 5 + 12, 19 – 2, 14 – 8
• Baseline maths test (instructed to perform similar)
5/04/12
9
Study 2 – Dual Task
Design • 6 x 6 balanced Latin square design • Series of planned comparisons with alpha adjusted
A B F C E D B C A D F E C D B E A F D E C F B A E F D A C B F A E B D C
Study 2 – Dual Task
Study 2 – Dual Task
Audio Condition
1 2 3 NC1
4 NC2
5 NC1
6 NC2
Single Task ✓
✓
✓
No Headphone ✓
✓
Dual Task ✓
✓
✓
NC Active
✓
✓
✓
✓
Brief through speaker
✓
✓
Brief through headphone
✓
✓
✓
✓
Wideband noise ✓
✓
✓
✓
✓
✓
5/04/12
10
Study 2 – Results
• No differences between headphones (single – 3 & 4 or dual task 5 & 6) largest F, F(1, 35) = 1.22, p = .276.
• Single task - NC (7.85) compared to no-NC (5.72), F(1, 35) = 14.93, p < .001.
• Dual task - NC (6.47) compared to no-NC (5.36), F(1, 35) = 13.18 , p = .001.
**Fewer questions answered in dual task than baseline No difference in error rate (baseline and dual task)**
Study 2 – Results
• NC (7.85) in single task compared to NC in dual task (6.47), F(1, 35) = 27.32, p < .001.
• NC (6.47) in dual task compared to single task but no headphones (5.72), F(1, 35) = 1.89, p = .184.
• Beneficial effects of NC are nullified if user elects to engage in 2nd task
• However, performance no worse than not using headphones (current situation in commercial aviation).
Study 3 – Masking
Aim 1. Examine the effects of masking and noise cancelling
headphones on task performance (commercial aviation)
Participants • 37 prior to screening, 36 (21 male) • Average age 21 years • Normal hearing (tested)
5/04/12
11
Study 3 – Masking
Design • Balanced Latin square design (6 x 6) • DV – correct # on fill-in-the-blanks test (max 12) • Aircraft noise 65 dB(A) • Audio briefs 70 dB(A) • Masking – low 50 dB(A) and loud music 70 dB(A)
Study 3 - Masking
Audio Condition
1 2 3
4
5 6
No Headphone ✓
Passive Headphones
✓ ✓
NC Active ✓ ✓ ✓
Low Music ✓ ✓
High Music ✓ ✓
Brief through speaker
✓
✓
✓
✓ ✓
Brief through headphone
✓
Wideband noise ✓
✓
✓
✓
✓ ✓
Study 3 – Results
• NC (7.03) better than no-headphones (6.06). • Passive headphones, low music (4.42) better than high
music masking (1.31). • Active headphones, low music (5.83) better than high
music masking (2.00).
5/04/12
12
Study 3 – Results
• Low masking – Active NC (5.83) better than Passive NC (4.42).
• High masking – Passive vs Active. No difference.
• No headphones vs Active low mask - No difference.
Study 3 – Results
• Active NC better than passive headphones.
• Active NC with low masking (music) no different to not
using headphones (current situation in commercial aviation).
All - Summary
• 3 separate studies • NC headphones repeatedly better than no headphones • No difference – NC dual task vs no headphones • No difference – NC masking low vs no headphones
• Noise impairs performance • Increasing the signal to noise ratio improves
performance
5/04/12
13
Summary
• Last Point - Increasing the signal to noise ratio improves performance
• Does that mean increase the volume of the audio brief?
Applied Perspective
The challenges for airlines: • Does a non transmitting PED interfere with sensitive
onboard electronic equipment? • How do you prevent passengers from listening to
‘loud’ music during the pre-flight safety brief?
In isolation, research suggests: • current practice is not best practice.
Future Research
Active noise reduction and: • Native language, • Age, • Different environments (older or newer aircraft), • Visual stimuli plus auditory stimuli, and • Fatigue.