Auditory alarm annoyance plagues clinicians, which results in alarms desensitization and ultimately affects patient care. Contributing to this problem are the International Electrotechnical Commission (IEC) 60601-1-8 alarms, a standardized set of melodic alarms used to convey information to clinicians in intensive care units. By design, IEC alarms employ flat amplitude (i.e., amplitude invariant) envelopes and are not reflective of naturally occurring sounds with percussive amplitude (i.e., decaying) envelopes. We present a series of three experiments evaluating the effect of amplitude envelope manipulation (i.e., incorporating percussive envelope) on memory and annoyance in IEC alarms synthesized using pure tones (experiment 1), complex tones (experiment 2) and assessing annoyance pre and post memory assessment (experiment 3). For the memory assessment, participants were assigned to learn either the flat alarms or percussive alarms. During the memory assessment, participants were informed of the alarm–referent pairings (study phase), practised identifying alarms (training phase), had a short break, and tested on their ability to identify alarms (evaluation phase). The annoyance assessment was a two alternative forced choice task where participants identified which alarm they perceived to be more annoying from a pair of alarms differing in either envelope-type or alarm-type. Across all experiments there was no difference in alarm learnability between those learning either flat or percussive alarms during the memory assessment. Annoyance assessments revealed that all participants chose the flat alarms to be more annoying than the percussive alarms, independent of the memory assessment condition. These results showcase the potential of using percussive alarms to reduce alarm annoyance without harming learnability, a cost-efficient manipulation. / Thesis / Master of Science (MSc) / Alarms in intensive care units are perceived as annoying, ultimately negatively affecting both clinicians and patients. These alarms are mandated by the International Electrotechnical Commission to have sustained or flat amplitude envelopes (i.e., referring to the change in loudness over time), which does not reflect naturally occurring stimuli that typically have decaying or percussive amplitude envelopes. The current experiments assessed the effect of percussive envelopes on alarm learnability and annoyance. We showed in a series of experiments that there is no difference in learning alarms with flat or percussive envelopes. However, we showed that alarms with percussive envelopes are perceived to be less annoying than alarms with flat envelopes. These results offer one potential solution to reduce alarm annoyance in intensive care units without harming the learnability of these alarms.

This thesis is composed of two independent manuscripts for publication. The first (Chapter 2) “More Detectable, Less Annoying. The Role of Temporal Variation in Envelope and Spectral Content on Detection and Annoyance” will be submitted to Psychological Science. This manuscript explores how detection and annoyance of sound are affected by temporal variation in two acoustic parameters; amplitude envelope and spectral content. The second (Chapter 3) “Improving Detectability of Auditory Interfaces Through Temporal Variation in Envelope” will be submitted to Human Factors. Here I build off of the previous manuscript by investigating tone detection in a split attention task more pertinent to the normative use of auditory interfaces. The author of this thesis is the primary author of both papers. / Thesis / Master of Science (MSc)

No description available.